GstInference GStreamer pipelines on PC

From RidgeRun Developer Wiki




Previous: Example pipelines Index Next: Example pipelines/NANO






Problems running the pipelines shown on this page? Please see our GStreamer Debugging guide for help.

Images used for classification task

  • Cat image to classify using Inception, Mobilenet and Resnet
Cat.jpg

Tensorflow

Inceptionv1

Image file

IMAGE_FILE=cat.jpg
MODEL_LOCATION='graph_inceptionv1_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV1/Logits/Predictions/Reshape_1'
GST_DEBUG=inceptionv1:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv1 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:09.549749856 26945       0xaf9cf0 LOG              inceptionv1 gstinceptionv1.c:199:gst_inceptionv1_preprocess:<net> Preprocess
0:00:10.672917685 26945       0xaf9cf0 LOG              inceptionv1 gstinceptionv1.c:231:gst_inceptionv1_postprocess:<net> Postprocess
0:00:10.672976676 26945       0xaf9cf0 LOG              inceptionv1 gstinceptionv1.c:252:gst_inceptionv1_postprocess:<net> Highest probability is label 284 : (0,691864)
0:00:10.673064576 26945       0xaf9cf0 LOG              inceptionv1 gstinceptionv1.c:199:gst_inceptionv1_preprocess:<net> Preprocess
0:00:11.793890820 26945       0xaf9cf0 LOG              inceptionv1 gstinceptionv1.c:231:gst_inceptionv1_postprocess:<net> Postprocess
0:00:11.793951581 26945       0xaf9cf0 LOG              inceptionv1 gstinceptionv1.c:252:gst_inceptionv1_postprocess:<net> Highest probability is label 284 : (0,691864)
0:00:11.794041207 26945       0xaf9cf0 LOG              inceptionv1 gstinceptionv1.c:199:gst_inceptionv1_preprocess:<net> Preprocess
0:00:12.920027410 26945       0xaf9cf0 LOG              inceptionv1 gstinceptionv1.c:231:gst_inceptionv1_postprocess:<net> Postprocess
0:00:12.920093762 26945       0xaf9cf0 LOG              inceptionv1 gstinceptionv1.c:252:gst_inceptionv1_postprocess:<net> Highest probability is label 284 : (0,691864)

Video file

VIDEO_FILE='cat.mp4'
MODEL_LOCATION='graph_inceptionv1_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV1/Logits/Predictions/Reshape_1'
GST_DEBUG=inceptionv1:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv1 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:11.878158663 27048      0x1d49800 LOG              inceptionv1 gstinceptionv1.c:199:gst_inceptionv1_preprocess:<net> Preprocess
0:00:13.006776924 27048      0x1d49800 LOG              inceptionv1 gstinceptionv1.c:231:gst_inceptionv1_postprocess:<net> Postprocess
0:00:13.006847113 27048      0x1d49800 LOG              inceptionv1 gstinceptionv1.c:252:gst_inceptionv1_postprocess:<net> Highest probability is label 282 : (0,594995)
0:00:13.006946305 27048      0x1d49800 LOG              inceptionv1 gstinceptionv1.c:199:gst_inceptionv1_preprocess:<net> Preprocess
0:00:14.170203673 27048      0x1d49800 LOG              inceptionv1 gstinceptionv1.c:231:gst_inceptionv1_postprocess:<net> Postprocess
0:00:14.170277808 27048      0x1d49800 LOG              inceptionv1 gstinceptionv1.c:252:gst_inceptionv1_postprocess:<net> Highest probability is label 282 : (0,595920)
0:00:14.170384768 27048      0x1d49800 LOG              inceptionv1 gstinceptionv1.c:199:gst_inceptionv1_preprocess:<net> Preprocess
0:00:15.285901546 27048      0x1d49800 LOG              inceptionv1 gstinceptionv1.c:231:gst_inceptionv1_postprocess:<net> Postprocess
0:00:15.285964794 27048      0x1d49800 LOG              inceptionv1 gstinceptionv1.c:252:gst_inceptionv1_postprocess:<net> Highest probability is label 282 : (0,593185)

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv1_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV1/Logits/Predictions/Reshape_1'
GST_DEBUG=inceptionv1:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv1 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:14.614862363 27227      0x19cd4a0 LOG              inceptionv1 gstinceptionv1.c:199:gst_inceptionv1_preprocess:<net> Preprocess
0:00:15.737842669 27227      0x19cd4a0 LOG              inceptionv1 gstinceptionv1.c:231:gst_inceptionv1_postprocess:<net> Postprocess
0:00:15.737912053 27227      0x19cd4a0 LOG              inceptionv1 gstinceptionv1.c:252:gst_inceptionv1_postprocess:<net> Highest probability is label 838 : (0,105199)
0:00:15.738007534 27227      0x19cd4a0 LOG              inceptionv1 gstinceptionv1.c:199:gst_inceptionv1_preprocess:<net> Preprocess
0:00:16.855603761 27227      0x19cd4a0 LOG              inceptionv1 gstinceptionv1.c:231:gst_inceptionv1_postprocess:<net> Postprocess
0:00:16.855673578 27227      0x19cd4a0 LOG              inceptionv1 gstinceptionv1.c:252:gst_inceptionv1_postprocess:<net> Highest probability is label 838 : (0,093981)
0:00:16.855768558 27227      0x19cd4a0 LOG              inceptionv1 gstinceptionv1.c:199:gst_inceptionv1_preprocess:<net> Preprocess
0:00:17.980784789 27227      0x19cd4a0 LOG              inceptionv1 gstinceptionv1.c:231:gst_inceptionv1_postprocess:<net> Postprocess
0:00:17.980849612 27227      0x19cd4a0 LOG              inceptionv1 gstinceptionv1.c:252:gst_inceptionv1_postprocess:<net> Highest probability is label 838 : (0,077824)

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv1_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV1/Logits/Predictions/Reshape_1'
LABELS='imagenet_labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
inceptionv1 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

Inceptionv2

Image file

IMAGE_FILE=cat.jpg
MODEL_LOCATION='graph_inceptionv2_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='Softmax'
GST_DEBUG=inceptionv2:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! videorate ! queue ! net.sink_model \
inceptionv2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:09.549749856 26945       0xaf9cf0 LOG              inceptionv2 gstinceptionv2.c:199:gst_inceptionv2_preprocess:<net> Preprocess
0:00:10.672917685 26945       0xaf9cf0 LOG              inceptionv2 gstinceptionv2.c:231:gst_inceptionv2_postprocess:<net> Postprocess
0:00:10.672976676 26945       0xaf9cf0 LOG              inceptionv2 gstinceptionv2.c:252:gst_inceptionv2_postprocess:<net> Highest probability is label 284 : (0,691864)
0:00:10.673064576 26945       0xaf9cf0 LOG              inceptionv2 gstinceptionv2.c:199:gst_inceptionv2_preprocess:<net> Preprocess
0:00:11.793890820 26945       0xaf9cf0 LOG              inceptionv2 gstinceptionv2.c:231:gst_inceptionv2_postprocess:<net> Postprocess
0:00:11.793951581 26945       0xaf9cf0 LOG              inceptionv2 gstinceptionv2.c:252:gst_inceptionv2_postprocess:<net> Highest probability is label 284 : (0,691864)
0:00:11.794041207 26945       0xaf9cf0 LOG              inceptionv2 gstinceptionv2.c:199:gst_inceptionv2_preprocess:<net> Preprocess
0:00:12.920027410 26945       0xaf9cf0 LOG              inceptionv2 gstinceptionv2.c:231:gst_inceptionv2_postprocess:<net> Postprocess
0:00:12.920093762 26945       0xaf9cf0 LOG              inceptionv2 gstinceptionv2.c:252:gst_inceptionv2_postprocess:<net> Highest probability is label 284 : (0,691864)

Video file

VIDEO_FILE='cat.mp4'
MODEL_LOCATION='graph_inceptionv2_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='Softmax'
GST_DEBUG=inceptionv2:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:11.878158663 27048      0x1d49800 LOG              inceptionv2 gstinceptionv2.c:199:gst_inceptionv2_preprocess:<net> Preprocess
0:00:13.006776924 27048      0x1d49800 LOG              inceptionv2 gstinceptionv2.c:231:gst_inceptionv2_postprocess:<net> Postprocess
0:00:13.006847113 27048      0x1d49800 LOG              inceptionv2 gstinceptionv2.c:252:gst_inceptionv2_postprocess:<net> Highest probability is label 282 : (0,594995)
0:00:13.006946305 27048      0x1d49800 LOG              inceptionv2 gstinceptionv2.c:199:gst_inceptionv2_preprocess:<net> Preprocess
0:00:14.170203673 27048      0x1d49800 LOG              inceptionv2 gstinceptionv2.c:231:gst_inceptionv2_postprocess:<net> Postprocess
0:00:14.170277808 27048      0x1d49800 LOG              inceptionv2 gstinceptionv2.c:252:gst_inceptionv2_postprocess:<net> Highest probability is label 282 : (0,595920)
0:00:14.170384768 27048      0x1d49800 LOG              inceptionv2 gstinceptionv2.c:199:gst_inceptionv2_preprocess:<net> Preprocess
0:00:15.285901546 27048      0x1d49800 LOG              inceptionv2 gstinceptionv2.c:231:gst_inceptionv2_postprocess:<net> Postprocess
0:00:15.285964794 27048      0x1d49800 LOG              inceptionv2 gstinceptionv2.c:252:gst_inceptionv2_postprocess:<net> Highest probability is label 282 : (0,593185)

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv2_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='Softmax'
GST_DEBUG=inceptionv2:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:14.614862363 27227      0x19cd4a0 LOG              inceptionv2 gstinceptionv2.c:199:gst_inceptionv2_preprocess:<net> Preprocess
0:00:15.737842669 27227      0x19cd4a0 LOG              inceptionv2 gstinceptionv2.c:231:gst_inceptionv2_postprocess:<net> Postprocess
0:00:15.737912053 27227      0x19cd4a0 LOG              inceptionv2 gstinceptionv2.c:252:gst_inceptionv2_postprocess:<net> Highest probability is label 838 : (0,105199)
0:00:15.738007534 27227      0x19cd4a0 LOG              inceptionv2 gstinceptionv2.c:199:gst_inceptionv2_preprocess:<net> Preprocess
0:00:16.855603761 27227      0x19cd4a0 LOG              inceptionv2 gstinceptionv2.c:231:gst_inceptionv2_postprocess:<net> Postprocess
0:00:16.855673578 27227      0x19cd4a0 LOG              inceptionv2 gstinceptionv2.c:252:gst_inceptionv2_postprocess:<net> Highest probability is label 838 : (0,093981)
0:00:16.855768558 27227      0x19cd4a0 LOG              inceptionv2 gstinceptionv2.c:199:gst_inceptionv2_preprocess:<net> Preprocess
0:00:17.980784789 27227      0x19cd4a0 LOG              inceptionv2 gstinceptionv2.c:231:gst_inceptionv2_postprocess:<net> Postprocess
0:00:17.980849612 27227      0x19cd4a0 LOG              inceptionv2 gstinceptionv2.c:252:gst_inceptionv2_postprocess:<net> Highest probability is label 838 : (0,077824)

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv2_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='Softmax'
LABELS='imagenet_labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
inceptionv2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

Inceptionv3

Image file

IMAGE_FILE=cat.jpg
MODEL_LOCATION='graph_inceptionv3_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV3/Predictions/Reshape_1
GST_DEBUG=inceptionv3:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv3 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:09.549749856 26945       0xaf9cf0 LOG              inceptionv3 gstinceptionv3.c:199:gst_inceptionv3_preprocess:<net> Preprocess
0:00:10.672917685 26945       0xaf9cf0 LOG              inceptionv3 gstinceptionv3.c:231:gst_inceptionv3_postprocess:<net> Postprocess
0:00:10.672976676 26945       0xaf9cf0 LOG              inceptionv3 gstinceptionv3.c:252:gst_inceptionv3_postprocess:<net> Highest probability is label 284 : (0,691864)
0:00:10.673064576 26945       0xaf9cf0 LOG              inceptionv3 gstinceptionv3.c:199:gst_inceptionv3_preprocess:<net> Preprocess
0:00:11.793890820 26945       0xaf9cf0 LOG              inceptionv3 gstinceptionv3.c:231:gst_inceptionv3_postprocess:<net> Postprocess
0:00:11.793951581 26945       0xaf9cf0 LOG              inceptionv3 gstinceptionv3.c:252:gst_inceptionv3_postprocess:<net> Highest probability is label 284 : (0,691864)
0:00:11.794041207 26945       0xaf9cf0 LOG              inceptionv3 gstinceptionv3.c:199:gst_inceptionv3_preprocess:<net> Preprocess
0:00:12.920027410 26945       0xaf9cf0 LOG              inceptionv3 gstinceptionv3.c:231:gst_inceptionv3_postprocess:<net> Postprocess
0:00:12.920093762 26945       0xaf9cf0 LOG              inceptionv3 gstinceptionv3.c:252:gst_inceptionv3_postprocess:<net> Highest probability is label 284 : (0,691864)

Video file

VIDEO_FILE='cat.mp4'
MODEL_LOCATION='graph_inceptionv3_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV3/Predictions/Reshape_1
GST_DEBUG=inceptionv3:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv3 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:11.878158663 27048      0x1d49800 LOG              inceptionv3 gstinceptionv3.c:199:gst_inceptionv3_preprocess:<net> Preprocess
0:00:13.006776924 27048      0x1d49800 LOG              inceptionv3 gstinceptionv3.c:231:gst_inceptionv3_postprocess:<net> Postprocess
0:00:13.006847113 27048      0x1d49800 LOG              inceptionv3 gstinceptionv3.c:252:gst_inceptionv3_postprocess:<net> Highest probability is label 282 : (0,594995)
0:00:13.006946305 27048      0x1d49800 LOG              inceptionv3 gstinceptionv3.c:199:gst_inceptionv3_preprocess:<net> Preprocess
0:00:14.170203673 27048      0x1d49800 LOG              inceptionv3 gstinceptionv3.c:231:gst_inceptionv3_postprocess:<net> Postprocess
0:00:14.170277808 27048      0x1d49800 LOG              inceptionv3 gstinceptionv3.c:252:gst_inceptionv3_postprocess:<net> Highest probability is label 282 : (0,595920)
0:00:14.170384768 27048      0x1d49800 LOG              inceptionv3 gstinceptionv3.c:199:gst_inceptionv3_preprocess:<net> Preprocess
0:00:15.285901546 27048      0x1d49800 LOG              inceptionv3 gstinceptionv3.c:231:gst_inceptionv3_postprocess:<net> Postprocess
0:00:15.285964794 27048      0x1d49800 LOG              inceptionv3 gstinceptionv3.c:252:gst_inceptionv3_postprocess:<net> Highest probability is label 282 : (0,593185)

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv3_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV3/Predictions/Reshape_1'
GST_DEBUG=inceptionv3:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv3 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:14.614862363 27227      0x19cd4a0 LOG              inceptionv3 gstinceptionv3.c:199:gst_inceptionv3_preprocess:<net> Preprocess
0:00:15.737842669 27227      0x19cd4a0 LOG              inceptionv3 gstinceptionv3.c:231:gst_inceptionv3_postprocess:<net> Postprocess
0:00:15.737912053 27227      0x19cd4a0 LOG              inceptionv3 gstinceptionv3.c:252:gst_inceptionv3_postprocess:<net> Highest probability is label 838 : (0,105199)
0:00:15.738007534 27227      0x19cd4a0 LOG              inceptionv3 gstinceptionv3.c:199:gst_inceptionv3_preprocess:<net> Preprocess
0:00:16.855603761 27227      0x19cd4a0 LOG              inceptionv3 gstinceptionv3.c:231:gst_inceptionv3_postprocess:<net> Postprocess
0:00:16.855673578 27227      0x19cd4a0 LOG              inceptionv3 gstinceptionv3.c:252:gst_inceptionv3_postprocess:<net> Highest probability is label 838 : (0,093981)
0:00:16.855768558 27227      0x19cd4a0 LOG              inceptionv3 gstinceptionv3.c:199:gst_inceptionv3_preprocess:<net> Preprocess
0:00:17.980784789 27227      0x19cd4a0 LOG              inceptionv3 gstinceptionv3.c:231:gst_inceptionv3_postprocess:<net> Postprocess
0:00:17.980849612 27227      0x19cd4a0 LOG              inceptionv3 gstinceptionv3.c:252:gst_inceptionv3_postprocess:<net> Highest probability is label 838 : (0,077824)

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv3_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV3/Predictions/Reshape_1'
LABELS='imagenet_labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
inceptionv3 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

Inceptionv4

Image file

IMAGE_FILE=cat.jpg
MODEL_LOCATION='graph_inceptionv4_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! videorate ! queue ! net.sink_model \
inceptionv4 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:09.549749856 26945       0xaf9cf0 LOG              inceptionv4 gstinceptionv4.c:199:gst_inceptionv4_preprocess:<net> Preprocess
0:00:10.672917685 26945       0xaf9cf0 LOG              inceptionv4 gstinceptionv4.c:231:gst_inceptionv4_postprocess:<net> Postprocess
0:00:10.672976676 26945       0xaf9cf0 LOG              inceptionv4 gstinceptionv4.c:252:gst_inceptionv4_postprocess:<net> Highest probability is label 284 : (0,691864)
0:00:10.673064576 26945       0xaf9cf0 LOG              inceptionv4 gstinceptionv4.c:199:gst_inceptionv4_preprocess:<net> Preprocess
0:00:11.793890820 26945       0xaf9cf0 LOG              inceptionv4 gstinceptionv4.c:231:gst_inceptionv4_postprocess:<net> Postprocess
0:00:11.793951581 26945       0xaf9cf0 LOG              inceptionv4 gstinceptionv4.c:252:gst_inceptionv4_postprocess:<net> Highest probability is label 284 : (0,691864)
0:00:11.794041207 26945       0xaf9cf0 LOG              inceptionv4 gstinceptionv4.c:199:gst_inceptionv4_preprocess:<net> Preprocess
0:00:12.920027410 26945       0xaf9cf0 LOG              inceptionv4 gstinceptionv4.c:231:gst_inceptionv4_postprocess:<net> Postprocess
0:00:12.920093762 26945       0xaf9cf0 LOG              inceptionv4 gstinceptionv4.c:252:gst_inceptionv4_postprocess:<net> Highest probability is label 284 : (0,691864)

Video file

VIDEO_FILE='cat.mp4'
MODEL_LOCATION='graph_inceptionv4_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv4 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:11.878158663 27048      0x1d49800 LOG              inceptionv4 gstinceptionv4.c:199:gst_inceptionv4_preprocess:<net> Preprocess
0:00:13.006776924 27048      0x1d49800 LOG              inceptionv4 gstinceptionv4.c:231:gst_inceptionv4_postprocess:<net> Postprocess
0:00:13.006847113 27048      0x1d49800 LOG              inceptionv4 gstinceptionv4.c:252:gst_inceptionv4_postprocess:<net> Highest probability is label 282 : (0,594995)
0:00:13.006946305 27048      0x1d49800 LOG              inceptionv4 gstinceptionv4.c:199:gst_inceptionv4_preprocess:<net> Preprocess
0:00:14.170203673 27048      0x1d49800 LOG              inceptionv4 gstinceptionv4.c:231:gst_inceptionv4_postprocess:<net> Postprocess
0:00:14.170277808 27048      0x1d49800 LOG              inceptionv4 gstinceptionv4.c:252:gst_inceptionv4_postprocess:<net> Highest probability is label 282 : (0,595920)
0:00:14.170384768 27048      0x1d49800 LOG              inceptionv4 gstinceptionv4.c:199:gst_inceptionv4_preprocess:<net> Preprocess
0:00:15.285901546 27048      0x1d49800 LOG              inceptionv4 gstinceptionv4.c:231:gst_inceptionv4_postprocess:<net> Postprocess
0:00:15.285964794 27048      0x1d49800 LOG              inceptionv4 gstinceptionv4.c:252:gst_inceptionv4_postprocess:<net> Highest probability is label 282 : (0,593185)

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv4_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv4 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:14.614862363 27227      0x19cd4a0 LOG              inceptionv4 gstinceptionv4.c:199:gst_inceptionv4_preprocess:<net> Preprocess
0:00:15.737842669 27227      0x19cd4a0 LOG              inceptionv4 gstinceptionv4.c:231:gst_inceptionv4_postprocess:<net> Postprocess
0:00:15.737912053 27227      0x19cd4a0 LOG              inceptionv4 gstinceptionv4.c:252:gst_inceptionv4_postprocess:<net> Highest probability is label 838 : (0,105199)
0:00:15.738007534 27227      0x19cd4a0 LOG              inceptionv4 gstinceptionv4.c:199:gst_inceptionv4_preprocess:<net> Preprocess
0:00:16.855603761 27227      0x19cd4a0 LOG              inceptionv4 gstinceptionv4.c:231:gst_inceptionv4_postprocess:<net> Postprocess
0:00:16.855673578 27227      0x19cd4a0 LOG              inceptionv4 gstinceptionv4.c:252:gst_inceptionv4_postprocess:<net> Highest probability is label 838 : (0,093981)
0:00:16.855768558 27227      0x19cd4a0 LOG              inceptionv4 gstinceptionv4.c:199:gst_inceptionv4_preprocess:<net> Preprocess
0:00:17.980784789 27227      0x19cd4a0 LOG              inceptionv4 gstinceptionv4.c:231:gst_inceptionv4_postprocess:<net> Postprocess
0:00:17.980849612 27227      0x19cd4a0 LOG              inceptionv4 gstinceptionv4.c:252:gst_inceptionv4_postprocess:<net> Highest probability is label 838 : (0,077824)

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv4_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
LABELS='imagenet_labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
inceptionv4 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

MobileNetv2

Image file

IMAGE_FILE=cat.jpg
MODEL_LOCATION='graph_mobilenetv2_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='MobilenetV2/Predictions/Reshape_1'
GST_DEBUG=mobilenetv2:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! queue ! net.sink_model \
mobilenetv2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:01.660006560    18      0x1138d90 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:02.387938090    18      0x1138d90 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:02.387975769    18      0x1138d90 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 282 : (0.183014)
0:00:02.390193061    18      0x1138d90 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:02.436405691    18      0x1138d90 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:02.436427612    18      0x1138d90 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 282 : (0.183014)
0:00:02.437487341    18      0x1138d90 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:02.467100635    18      0x1138d90 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:02.467123380    18      0x1138d90 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 282 : (0.183014)
0:00:02.468190400    18      0x1138d90 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:02.497410196    18      0x1138d90 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:02.497432133    18      0x1138d90 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 282 : (0.183014)

Video file

VIDEO_FILE='cat.mp4'
MODEL_LOCATION='graph_mobilenetv2_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='MobilenetV2/Predictions/Reshape_1'
GST_DEBUG=mobilenetv2:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
mobilenetv2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:01.901025239    60      0x248dc00 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:02.176679623    60      0x248dc00 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:02.176702018    60      0x248dc00 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0.306512)
0:00:02.176740543    60      0x248dc00 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:02.208491216    60      0x248dc00 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:02.208517379    60      0x248dc00 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0.307123)
0:00:02.208559346    60      0x248dc00 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:02.238110702    60      0x248dc00 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:02.238133192    60      0x248dc00 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0.318610)
0:00:02.238168437    60      0x248dc00 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:02.267137242    60      0x248dc00 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:02.267159969    60      0x248dc00 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0.323910)
0

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_mobilenetv2_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='MobilenetV2/Predictions/Reshape_1'
GST_DEBUG=mobilenetv2:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
mobilenetv2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:01.177456974   114      0x2db18a0 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:01.415812954   114      0x2db18a0 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:01.415834549   114      0x2db18a0 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 814 : (0.056321)
0:00:01.415870129   114      0x2db18a0 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:01.447472786   114      0x2db18a0 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:01.447492954   114      0x2db18a0 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 814 : (0.093839)
0:00:01.447522930   114      0x2db18a0 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:01.477011889   114      0x2db18a0 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:01.477031365   114      0x2db18a0 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 814 : (0.114949)
0:00:01.477061599   114      0x2db18a0 LOG              mobilenetv2 gstmobilenetv2.c:140:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:01.506820855   114      0x2db18a0 LOG              mobilenetv2 gstmobilenetv2.c:151:gst_mobilenetv2_postprocess:<net> Postprocess
0:00:01.506841456   114      0x2db18a0 LOG              mobilenetv2 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 814 : (0.097499)

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_mobilenetv2_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='MobilenetV2/Predictions/Reshape_1'
LABELS='imagenet_labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
mobilenetv2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

Resnet50v1

Image file

IMAGE_FILE=cat.jpg
MODEL_LOCATION='graph_resnetv1_tensorflow.pb'
INPUT_LAYER='input_tensor'
OUTPUT_LAYER='softmax_tensor'
GST_DEBUG=resnet50v1:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! queue ! net.sink_model \
resnet50v1 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:01.944768522   157       0xfccd90 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:01.944803563   157       0xfccd90 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 284 : (0.271051)
0:00:01.947003178   157       0xfccd90 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:02.111978575   157       0xfccd90 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:02.112000558   157       0xfccd90 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 284 : (0.271051)
0:00:02.113091931   157       0xfccd90 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:02.212289668   157       0xfccd90 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:02.212310188   157       0xfccd90 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 284 : (0.271051)

Video file

VIDEO_FILE='cat.mp4'
MODEL_LOCATION='graph_resnetv1_tensorflow.pb'
INPUT_LAYER='input_tensor'
OUTPUT_LAYER='softmax_tensor'
GST_DEBUG=resnet50v1:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
resnet50v1 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:00.915688134   240      0x18cbc00 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:00.915709354   240      0x18cbc00 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 283 : (0.537144)
0:00:00.915747394   240      0x18cbc00 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:01.018904132   240      0x18cbc00 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:01.018924929   240      0x18cbc00 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 283 : (0.538948)
0:00:01.018976948   240      0x18cbc00 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:01.120286331   240      0x18cbc00 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:01.120306927   240      0x18cbc00 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 283 : (0.525331)

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_resnetv1_tensorflow.pb'
INPUT_LAYER='input_tensor'
OUTPUT_LAYER='softmax_tensor'
GST_DEBUG=resnet50v1:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
resnet50v1 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:01.842896607   294      0x14dd8a0 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:01.842917966   294      0x14dd8a0 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 425 : (0.048243)
0:00:01.842955409   294      0x14dd8a0 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:01.948003304   294      0x14dd8a0 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:01.948024035   294      0x14dd8a0 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 611 : (0.065279)
0:00:01.948055304   294      0x14dd8a0 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:02.052442770   294      0x14dd8a0 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:02.052463202   294      0x14dd8a0 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 611 : (0.089816)

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_resnetv1_tensorflow.pb'
INPUT_LAYER='input_tensor'
OUTPUT_LAYER='softmax_tensor'
LABELS='imagenet_labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
resnet50v1 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

TinyYolov2

Image file

  • Get the graph used on this example from RidgeRun Store
  • You will need an image file from one of TinyYOLO classes
  • Pipeline
IMAGE_FILE='cat.jpg'
MODEL_LOCATION='graph_tinyyolov2_tensorflow.pb'
INPUT_LAYER='input/Placeholder'
OUTPUT_LAYER='add_8'
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! videorate ! queue ! net.sink_model \
tinyyolov2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:06.401015400 12340      0x1317cf0 LOG               tinyyolov2 gsttinyyolov2.c:479:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:06.817243785 12340      0x1317cf0 LOG               tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:06.817315935 12340      0x1317cf0 LOG               tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:7, x:-55,170727, y:25,507316, width:396,182867, height:423,241143, prob:14,526398]
0:00:06.817426814 12340      0x1317cf0 LOG               tinyyolov2 gsttinyyolov2.c:479:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:07.236310555 12340      0x1317cf0 LOG               tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:07.236379100 12340      0x1317cf0 LOG               tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:7, x:-55,170727, y:25,507316, width:396,182867, height:423,241143, prob:14,526398]
0:00:07.236486242 12340      0x1317cf0 LOG               tinyyolov2 gsttinyyolov2.c:479:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:07.659870194 12340      0x1317cf0 LOG               tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:07.659942388 12340      0x1317cf0 LOG               tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:7, x:-55,170727, y:25,507316, width:396,182867, height:423,241143, prob:14,526398]

Video file

  • Get the graph used on this example from RidgeRun Store
  • You will need a video file from one of TinyYOLO classes
  • Pipeline
VIDEO_FILE='cat.mp4'
MODEL_LOCATION='graph_tinyyolov2_tensorflow.pb'
INPUT_LAYER='input/Placeholder'
OUTPUT_LAYER='add_8'
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
tinyyolov2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:08.545063684 12504       0xce4400 LOG               tinyyolov2 gsttinyyolov2.c:479:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:08.955522899 12504       0xce4400 LOG               tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:08.955600820 12504       0xce4400 LOG               tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:7, x:-36,012765, y:-37,118160, width:426,351621, height:480,353663, prob:14,378592]
0:00:08.955824676 12504       0xce4400 LOG               tinyyolov2 gsttinyyolov2.c:479:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:09.364908234 12504       0xce4400 LOG               tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:09.364970901 12504       0xce4400 LOG               tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:7, x:-36,490694, y:-38,108817, width:427,474399, height:482,318385, prob:14,257683]
0:00:09.365090340 12504       0xce4400 LOG               tinyyolov2 gsttinyyolov2.c:479:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:09.775848590 12504       0xce4400 LOG               tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:09.775932404 12504       0xce4400 LOG               tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:7, x:-35,991940, y:-37,482425, width:426,533537, height:480,917142, prob:14,313076]

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_tinyyolov2_tensorflow.pb'
INPUT_LAYER='input/Placeholder'
OUTPUT_LAYER='add_8'
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
tinyyolov2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:06.823064776 12678       0xec24a0 LOG               tinyyolov2 gsttinyyolov2.c:479:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:07.242114002 12678       0xec24a0 LOG               tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:07.242183276 12678       0xec24a0 LOG               tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:14, x:116,796387, y:-31,424289, width:240,876587, height:536,305261, prob:11,859128]
0:00:07.242293677 12678       0xec24a0 LOG               tinyyolov2 gsttinyyolov2.c:479:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:07.660324555 12678       0xec24a0 LOG               tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:07.660388215 12678       0xec24a0 LOG               tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:14, x:113,453324, y:-27,681194, width:248,010337, height:528,964842, prob:11,603928]
0:00:07.660503502 12678       0xec24a0 LOG               tinyyolov2 gsttinyyolov2.c:479:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:08.079154860 12678       0xec24a0 LOG               tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:08.079230404 12678       0xec24a0 LOG               tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:14, x:113,736444, y:-33,747251, width:246,987389, height:541,188374, prob:11,888664]

Visualization with detection overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_tinyyolov2_tensorflow.pb'
INPUT_LAYER='input/Placeholder'
OUTPUT_LAYER='add_8'
LABELS='labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
tinyyolov2 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER backend::output-layer=$OUTPUT_LAYER \
net.src_bypass ! detectionoverlay labels="$(cat $LABELS)" font-scale=1 thickness=2 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output


TinyYolov3

Image file

  • Get the graph used on this example from RidgeRun Store
  • You will need an image file from one of TinyYOLO classes
  • Pipeline
IMAGE_FILE='cat.jpg'
MODEL_LOCATION='graph_tinyyolov3_tensorflow.pb'
INPUT_LAYER='inputs'
OUTPUT_LAYER='output_boxes'
GST_DEBUG=tinyyolov3:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! videorate ! queue ! net.sink_model \
tinyyolov3 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:06.401015400 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:06.817243785 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:06.817315935 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-55,170727, y:25,507316, width:396,182867, height:423,241143, prob:14,526398]
0:00:06.817426814 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:07.236310555 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:07.236379100 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-55,170727, y:25,507316, width:396,182867, height:423,241143, prob:14,526398]
0:00:07.236486242 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:07.659870194 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:07.659942388 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-55,170727, y:25,507316, width:396,182867, height:423,241143, prob:14,526398]

Video file

  • Get the graph used on this example from RidgeRun Store
  • You will need a video file from one of TinyYOLO classes
  • Pipeline
VIDEO_FILE='cat.mp4'
MODEL_LOCATION='graph_tinyyolov3_tensorflow.pb'
INPUT_LAYER='inputs'
OUTPUT_LAYER='output_boxes'
GST_DEBUG=tinyyolov3:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
tinyyolov3 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:08.545063684 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:08.955522899 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:08.955600820 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-36,012765, y:-37,118160, width:426,351621, height:480,353663, prob:14,378592]
0:00:08.955824676 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:09.364908234 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:09.364970901 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-36,490694, y:-38,108817, width:427,474399, height:482,318385, prob:14,257683]
0:00:09.365090340 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:09.775848590 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:09.775932404 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-35,991940, y:-37,482425, width:426,533537, height:480,917142, prob:14,313076]

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_tinyyolov3_tensorflow.pb'
INPUT_LAYER='inputs'
OUTPUT_LAYER='output_boxes'
GST_DEBUG=tinyyolov3:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
tinyyolov3 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER  backend::output-layer=$OUTPUT_LAYER
  • Output
0:00:06.823064776 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:07.242114002 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:07.242183276 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:14, x:116,796387, y:-31,424289, width:240,876587, height:536,305261, prob:11,859128]
0:00:07.242293677 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:07.660324555 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:07.660388215 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:14, x:113,453324, y:-27,681194, width:248,010337, height:528,964842, prob:11,603928]
0:00:07.660503502 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:08.079154860 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:08.079230404 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:14, x:113,736444, y:-33,747251, width:246,987389, height:541,188374, prob:11,888664]

Visualization with detection overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_tinyyolov3_tensorflow.pb'
INPUT_LAYER='inputs'
OUTPUT_LAYER='output_boxes'
LABELS='labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
tinyyolov3 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER backend::output-layer=$OUTPUT_LAYER \
net.src_bypass ! detectionoverlay labels="$(cat $LABELS)" font-scale=1 thickness=2 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

FaceNet

Visualization with embedding overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
  • LABELS and EMBEDDINGS files are in $PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings.
CAMERA='/dev/video0'
MODEL_LOCATION='graph_facenetv1_tensorflow.pb'
INPUT_LAYER='input'
OUTPUT_LAYER='output'
LABELS='$PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings/labels.txt'
EMBEDDINGS='$PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings/embeddings.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
facenetv1 name=net model-location=$MODEL_LOCATION backend=tensorflow backend::input-layer=$INPUT_LAYER backend::output-layer=$OUTPUT_LAYER \
net.src_bypass ! embeddingoverlay labels="$(cat $LABELS)" embeddings="$(cat $EMBEDDINGS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Embedding overlay example output

Tensorflow Lite

Inceptionv1

Image file

IMAGE_FILE=cat.jpg
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv1.tflite'
GST_DEBUG=inceptionv1:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv1 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.820787752 14910 0x55fad0914ed0 LOG              inceptionv1 gstinceptionv1.c:162:gst_inceptionv1_postprocess_old:<net> Postprocess
0:00:02.820811267 14910 0x55fad0914ed0 LOG              inceptionv1 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 283 : (0,124103)
0:00:02.820816935 14910 0x55fad0914ed0 LOG              inceptionv1 gstinceptionv1.c:187:gst_inceptionv1_postprocess_new:<net> Postprocess Meta
0:00:02.820909931 14910 0x55fad0914ed0 LOG              inceptionv1 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 49,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 224
    height : 224
  },
  classes : [
    {
      Id : 98
      Class : 283
      Label : tiger cat
      Probability : 0,124103
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Video file

VIDEO_FILE='cat.mp4'
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv1.tflite'
GST_DEBUG=inceptionv1:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv1 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.660861495 16805 0x558f50081850 LOG              inceptionv1 gstinceptionv1.c:150:gst_inceptionv1_preprocess:<net> Preprocess
0:00:02.704949141 16805 0x558f50081850 LOG              inceptionv1 gstinceptionv1.c:162:gst_inceptionv1_postprocess_old:<net> Postprocess
0:00:02.704973078 16805 0x558f50081850 LOG              inceptionv1 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,421085)
0:00:02.704978817 16805 0x558f50081850 LOG              inceptionv1 gstinceptionv1.c:187:gst_inceptionv1_postprocess_new:<net> Postprocess Meta
0:00:02.705073055 16805 0x558f50081850 LOG              inceptionv1 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 47,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 224
    height : 224
  },
  classes : [
    {
      Id : 94
      Class : 286
      Label : Egyptian cat
      Probability : 0,421085
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv1.tflite'
GST_DEBUG=inceptionv1:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv1 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.820787752 14910 0x55fad0914ed0 LOG              inceptionv1 gstinceptionv1.c:162:gst_inceptionv1_postprocess_old:<net> Postprocess
0:00:02.820811267 14910 0x55fad0914ed0 LOG              inceptionv1 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 283 : (0,124103)
0:00:02.820816935 14910 0x55fad0914ed0 LOG              inceptionv1 gstinceptionv1.c:187:gst_inceptionv1_postprocess_new:<net> Postprocess Meta
0:00:02.820909931 14910 0x55fad0914ed0 LOG              inceptionv1 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 49,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 224
    height : 224
  },
  classes : [
    {
      Id : 98
      Class : 283
      Label : tiger cat
      Probability : 0,124103
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv1.tflite'
LABELS='labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
inceptionv1 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"   \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" style=2  font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

Inceptionv2

Image file

IMAGE_FILE=cat.jpg
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv2.tflite'
GST_DEBUG=inceptionv2:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! videorate ! queue ! net.sink_model \
inceptionv2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:07.851985949 14671 0x563426d5ded0 LOG              inceptionv2 gstinceptionv2.c:217:gst_inceptionv2_preprocess:<net> Preprocess
0:00:07.931498739 14671 0x563426d5ded0 LOG              inceptionv2 gstinceptionv2.c:229:gst_inceptionv2_postprocess_old:<net> Postprocess
0:00:07.931528235 14671 0x563426d5ded0 LOG              inceptionv2 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,782229)
0:00:07.931538163 14671 0x563426d5ded0 LOG              inceptionv2 gstinceptionv2.c:254:gst_inceptionv2_postprocess_new:<net> Postprocess Meta
0:00:07.931645047 14671 0x563426d5ded0 LOG              inceptionv2 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 108,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 224
    height : 224
  },
  classes : [
    {
      Id : 216
      Class : 286
      Label : Egyptian cat
      Probability : 0,782229
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Video file

VIDEO_FILE='cat.mp4'
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv2.tflite'
GST_DEBUG=inceptionv2:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.664634195 16873 0x561f1782b850 LOG              inceptionv2 gstinceptionv2.c:229:gst_inceptionv2_postprocess_old:<net> Postprocess
0:00:02.664702995 16873 0x561f1782b850 LOG              inceptionv2 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,620495)
0:00:02.664716316 16873 0x561f1782b850 LOG              inceptionv2 gstinceptionv2.c:254:gst_inceptionv2_postprocess_new:<net> Postprocess Meta
0:00:02.665002849 16873 0x561f1782b850 LOG              inceptionv2 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 32,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 224
    height : 224
  },
  classes : [
    {
      Id : 64
      Class : 286
      Label : Egyptian cat
      Probability : 0,620495
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv2.tflite'
GST_DEBUG=inceptionv2:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:07.851985949 14671 0x563426d5ded0 LOG              inceptionv2 gstinceptionv2.c:217:gst_inceptionv2_preprocess:<net> Preprocess
0:00:07.931498739 14671 0x563426d5ded0 LOG              inceptionv2 gstinceptionv2.c:229:gst_inceptionv2_postprocess_old:<net> Postprocess
0:00:07.931528235 14671 0x563426d5ded0 LOG              inceptionv2 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,782229)
0:00:07.931538163 14671 0x563426d5ded0 LOG              inceptionv2 gstinceptionv2.c:254:gst_inceptionv2_postprocess_new:<net> Postprocess Meta
0:00:07.931645047 14671 0x563426d5ded0 LOG              inceptionv2 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 108,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 224
    height : 224
  },
  classes : [
    {
      Id : 216
      Class : 286
      Label : Egyptian cat
      Probability : 0,782229
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv2.tflite'
LABELS='labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
inceptionv2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"   \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" style=2 font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

Inceptionv3

Image file

IMAGE_FILE=cat.jpg
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv3.tflite'
GST_DEBUG=inceptionv3:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv3 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.176072412 14946 0x557b199d4ed0 LOG              inceptionv3 gstinceptionv3.c:161:gst_inceptionv3_postprocess_old:<net> Postprocess
0:00:02.176098336 14946 0x557b199d4ed0 LOG              inceptionv3 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,883076)
0:00:02.176122466 14946 0x557b199d4ed0 LOG              inceptionv3 gstinceptionv3.c:186:gst_inceptionv3_postprocess_new:<net> Postprocess Meta
0:00:02.176226140 14946 0x557b199d4ed0 LOG              inceptionv3 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 11,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 299
    height : 299
  },
  classes : [
    {
      Id : 22
      Class : 286
      Label : Egyptian cat
      Probability : 0,883076
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Video file

VIDEO_FILE='cat.mp4'
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv3.tflite'
GST_DEBUG=inceptionv3:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv3 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.685245277 16898 0x55b256c93850 LOG              inceptionv3 gstinceptionv3.c:161:gst_inceptionv3_postprocess_old:<net> Postprocess
0:00:02.685292515 16898 0x55b256c93850 LOG              inceptionv3 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,972109)
0:00:02.685299510 16898 0x55b256c93850 LOG              inceptionv3 gstinceptionv3.c:186:gst_inceptionv3_postprocess_new:<net> Postprocess Meta
0:00:02.685411145 16898 0x55b256c93850 LOG              inceptionv3 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 12,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 299
    height : 299
  },
  classes : [
    {
      Id : 24
      Class : 286
      Label : Egyptian cat
      Probability : 0,972109
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv3.tflite'
GST_DEBUG=inceptionv3:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv3 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.176072412 14946 0x557b199d4ed0 LOG              inceptionv3 gstinceptionv3.c:161:gst_inceptionv3_postprocess_old:<net> Postprocess
0:00:02.176098336 14946 0x557b199d4ed0 LOG              inceptionv3 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,883076)
0:00:02.176122466 14946 0x557b199d4ed0 LOG              inceptionv3 gstinceptionv3.c:186:gst_inceptionv3_postprocess_new:<net> Postprocess Meta
0:00:02.176226140 14946 0x557b199d4ed0 LOG              inceptionv3 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 11,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 299
    height : 299
  },
  classes : [
    {
      Id : 22
      Class : 286
      Label : Egyptian cat
      Probability : 0,883076
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv3.tflite'
LABELS='labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
inceptionv3 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"  \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" style=2  font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

Inceptionv4

Image file

IMAGE_FILE=cat.jpg
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv4.tflite'
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! videorate ! queue ! net.sink_model \
inceptionv4 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.039483790 14972 0x55c45dd99ed0 LOG              inceptionv4 gstinceptionv4.c:209:gst_inceptionv4_preprocess:<net> Preprocess
0:00:02.382000009 14972 0x55c45dd99ed0 LOG              inceptionv4 gstinceptionv4.c:221:gst_inceptionv4_postprocess_old:<net> Postprocess
0:00:02.382024685 14972 0x55c45dd99ed0 LOG              inceptionv4 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,956486)
0:00:02.382030318 14972 0x55c45dd99ed0 LOG              inceptionv4 gstinceptionv4.c:246:gst_inceptionv4_postprocess_new:<net> Postprocess Meta
0:00:02.382154899 14972 0x55c45dd99ed0 LOG              inceptionv4 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 5,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 299
    height : 299
  },
  classes : [
    {
      Id : 10
      Class : 286
      Label : Egyptian cat
      Probability : 0,956486
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Video file

VIDEO_FILE='cat.mp4'
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv4.tflite'
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv4 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:06.223168928 16998 0x55a84d9ae850 LOG              inceptionv4 gstinceptionv4.c:221:gst_inceptionv4_postprocess_old:<net> Postprocess
0:00:06.223196388 16998 0x55a84d9ae850 LOG              inceptionv4 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,947655)
0:00:06.223202015 16998 0x55a84d9ae850 LOG              inceptionv4 gstinceptionv4.c:246:gst_inceptionv4_postprocess_new:<net> Postprocess Meta
0:00:06.223294500 16998 0x55a84d9ae850 LOG              inceptionv4 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 18,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 299
    height : 299
  },
  classes : [
    {
      Id : 36
      Class : 286
      Label : Egyptian cat
      Probability : 0,947655
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
LABELS='labels.txt'
MODEL_LOCATION='graph_inceptionv4.tflite'
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
inceptionv4 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.039483790 14972 0x55c45dd99ed0 LOG              inceptionv4 gstinceptionv4.c:209:gst_inceptionv4_preprocess:<net> Preprocess
0:00:02.382000009 14972 0x55c45dd99ed0 LOG              inceptionv4 gstinceptionv4.c:221:gst_inceptionv4_postprocess_old:<net> Postprocess
0:00:02.382024685 14972 0x55c45dd99ed0 LOG              inceptionv4 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,956486)
0:00:02.382030318 14972 0x55c45dd99ed0 LOG              inceptionv4 gstinceptionv4.c:246:gst_inceptionv4_postprocess_new:<net> Postprocess Meta
0:00:02.382154899 14972 0x55c45dd99ed0 LOG              inceptionv4 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 5,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 299
    height : 299
  },
  classes : [
    {
      Id : 10
      Class : 286
      Label : Egyptian cat
      Probability : 0,956486
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_inceptionv4.tflite'
LABELS='labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
inceptionv4 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"  \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" style=2  font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

MobileNetv2

Image file

IMAGE_FILE=cat.jpg
LABELS='labels.txt'
MODEL_LOCATION='graph_mobilenetv2.tflite'
GST_DEBUG=mobilenetv2:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! queue ! net.sink_model \
mobilenetv2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.898115510 15109 0x562a0b2e7ed0 LOG              mobilenetv2 gstmobilenetv2.c:148:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:03.001354267 15109 0x562a0b2e7ed0 LOG              mobilenetv2 gstmobilenetv2.c:160:gst_mobilenetv2_postprocess_old:<net> Postprocess
0:00:03.001449439 15109 0x562a0b2e7ed0 LOG              mobilenetv2 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,761563)
0:00:03.001456716 15109 0x562a0b2e7ed0 LOG              mobilenetv2 gstmobilenetv2.c:185:gst_mobilenetv2_postprocess_new:<net> Postprocess Meta
0:00:03.001575055 15109 0x562a0b2e7ed0 LOG              mobilenetv2 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 28,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 224
    height : 224
  },
  classes : [
    {
      Id : 56
      Class : 286
      Label : Egyptian cat
      Probability : 0,761563
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Video file

VIDEO_FILE='cat.mp4'
LABELS='labels.txt'
MODEL_LOCATION='graph_mobilenetv2.tflite'
GST_DEBUG=mobilenetv2:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
mobilenetv2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.228695071 17037 0x5556ecccf850 LOG              mobilenetv2 gstmobilenetv2.c:148:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:02.323426275 17037 0x5556ecccf850 LOG              mobilenetv2 gstmobilenetv2.c:160:gst_mobilenetv2_postprocess_old:<net> Postprocess
0:00:02.323468784 17037 0x5556ecccf850 LOG              mobilenetv2 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,640671)
0:00:02.323477263 17037 0x5556ecccf850 LOG              mobilenetv2 gstmobilenetv2.c:185:gst_mobilenetv2_postprocess_new:<net> Postprocess Meta
0:00:02.323611426 17037 0x5556ecccf850 LOG              mobilenetv2 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 21,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 224
    height : 224
  },
  classes : [
    {
      Id : 42
      Class : 286
      Label : Egyptian cat
      Probability : 0,640671
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
LABELS='labels.txt'
MODEL_LOCATION='graph_mobilenetv2.tflite'
GST_DEBUG=mobilenetv2:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
mobilenetv2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.898115510 15109 0x562a0b2e7ed0 LOG              mobilenetv2 gstmobilenetv2.c:148:gst_mobilenetv2_preprocess:<net> Preprocess
0:00:03.001354267 15109 0x562a0b2e7ed0 LOG              mobilenetv2 gstmobilenetv2.c:160:gst_mobilenetv2_postprocess_old:<net> Postprocess
0:00:03.001449439 15109 0x562a0b2e7ed0 LOG              mobilenetv2 gstinferencedebug.c:74:gst_inference_print_highest_probability:<net> Highest probability is label 286 : (0,761563)
0:00:03.001456716 15109 0x562a0b2e7ed0 LOG              mobilenetv2 gstmobilenetv2.c:185:gst_mobilenetv2_postprocess_new:<net> Postprocess Meta
0:00:03.001575055 15109 0x562a0b2e7ed0 LOG              mobilenetv2 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 28,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 224
    height : 224
  },
  classes : [
    {
      Id : 56
      Class : 286
      Label : Egyptian cat
      Probability : 0,761563
      Classes : 1001
    }, 
  ],
  predictions : [
    
  ]
}

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_mobilenetv2.tflite'
LABELS='labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
mobilenetv2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"  \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" style=2  font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output


Resnet50v1

Image file

IMAGE_FILE=cat.jpg
LABELS='labels.txt'
MODEL_LOCATION='graph_resnetv1.tflite'
GST_DEBUG=resnet50v1:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! queue ! net.sink_model \
resnet50v1 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:01.944768522   157       0xfccd90 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:01.944803563   157       0xfccd90 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 284 : (0.271051)
0:00:01.947003178   157       0xfccd90 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:02.111978575   157       0xfccd90 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:02.112000558   157       0xfccd90 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 284 : (0.271051)
0:00:02.113091931   157       0xfccd90 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:02.212289668   157       0xfccd90 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:02.212310188   157       0xfccd90 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 284 : (0.271051)

Video file

VIDEO_FILE='cat.mp4'
LABELS='labels.txt'
MODEL_LOCATION='graph_resnetv1.tflite'
GST_DEBUG=resnet50v1:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
resnet50v1 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:00.915688134   240      0x18cbc00 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:00.915709354   240      0x18cbc00 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 283 : (0.537144)
0:00:00.915747394   240      0x18cbc00 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:01.018904132   240      0x18cbc00 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:01.018924929   240      0x18cbc00 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 283 : (0.538948)
0:00:01.018976948   240      0x18cbc00 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:01.120286331   240      0x18cbc00 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:01.120306927   240      0x18cbc00 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 283 : (0.525331)

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
LABELS='labels.txt'
MODEL_LOCATION='graph_resnetv1.tflite'
GST_DEBUG=resnet50v1:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
resnet50v1 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:01.842896607   294      0x14dd8a0 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:01.842917966   294      0x14dd8a0 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 425 : (0.048243)
0:00:01.842955409   294      0x14dd8a0 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:01.948003304   294      0x14dd8a0 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:01.948024035   294      0x14dd8a0 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 611 : (0.065279)
0:00:01.948055304   294      0x14dd8a0 LOG               resnet50v1 gstresnet50v1.c:145:gst_resnet50v1_preprocess:<net> Preprocess
0:00:02.052442770   294      0x14dd8a0 LOG               resnet50v1 gstresnet50v1.c:157:gst_resnet50v1_postprocess:<net> Postprocess
0:00:02.052463202   294      0x14dd8a0 LOG               resnet50v1 gstinferencedebug.c:73:gst_inference_print_highest_probability:<net> Highest probability is label 611 : (0.089816)

Visualization with classification overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
LABELS='labels.txt'
MODEL_LOCATION='graph_resnetv1.tflite'
LABELS='labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
resnet50v1 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"  \
net.src_bypass ! classificationoverlay labels="$(cat $LABELS)" style=2  font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output

TinyYolov2

Image file

  • Get the graph used on this example from RidgeRun Store
  • You will need an image file from one of TinyYOLO classes
  • Pipeline
IMAGE_FILE='cat.jpg'
LABELS='labels.txt'
MODEL_LOCATION='graph_tinyyolov2.tflite'
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! videorate ! queue ! net.sink_model \
tinyyolov2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:17.446720448 19333 0x55892c0ae770 LOG               tinyyolov2 gsttinyyolov2.c:286:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:17.548641827 19333 0x55892c0ae770 LOG               tinyyolov2 gsttinyyolov2.c:325:gst_tinyyolov2_postprocess_old:<net> Postprocess
0:00:17.548692764 19333 0x55892c0ae770 LOG               tinyyolov2 gstinferencedebug.c:93:gst_inference_print_boxes:<net> Box: [class:14, x:93,758068, y:71,626141, width:218,740955, height:334,471067, prob:10,713037]
0:00:17.548699121 19333 0x55892c0ae770 LOG               tinyyolov2 gsttinyyolov2.c:359:gst_tinyyolov2_postprocess_new:<net> Postprocess Meta
0:00:17.548705539 19333 0x55892c0ae770 LOG               tinyyolov2 gsttinyyolov2.c:366:gst_tinyyolov2_postprocess_new:<net> Number of predictions: 1
0:00:17.548816856 19333 0x55892c0ae770 LOG               tinyyolov2 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 299,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 416
    height : 416
  },
  classes : [
    
  ],
  predictions : [
    {
      id : 300,
      enabled : True,
      bbox : {
        x : 93
        y : 71
        width : 218
        height : 334
      },
      classes : [
        {
          Id : 280
          Class : 14
          Label : person
          Probability : 10,713037
          Classes : 20
        }, 
      ],
      predictions : [
        
      ]
    }, 
  ]
}

Video file

  • Get the graph used on this example from RidgeRun Store
  • You will need a video file from one of TinyYOLO classes
  • Pipeline
VIDEO_FILE='cat.mp4'
LABELS='labels.txt'
MODEL_LOCATION='graph_tinyyolov2.tflite'
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
tinyyolov2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:03.946822351 19396 0x55fc3118d680 LOG               tinyyolov2 gsttinyyolov2.c:325:gst_tinyyolov2_postprocess_old:<net> Postprocess
0:00:03.946899445 19396 0x55fc3118d680 LOG               tinyyolov2 gstinferencedebug.c:93:gst_inference_print_boxes:<net> Box: [class:14, x:62,124242, y:121,697849, width:215,944135, height:290,148073, prob:13,969749]
0:00:03.946905463 19396 0x55fc3118d680 LOG               tinyyolov2 gsttinyyolov2.c:359:gst_tinyyolov2_postprocess_new:<net> Postprocess Meta
0:00:03.946912573 19396 0x55fc3118d680 LOG               tinyyolov2 gsttinyyolov2.c:366:gst_tinyyolov2_postprocess_new:<net> Number of predictions: 1
0:00:03.947079421 19396 0x55fc3118d680 LOG               tinyyolov2 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 58,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 416
    height : 416
  },
  classes : [
    
  ],
  predictions : [
    {
      id : 59,
      enabled : True,
      bbox : {
        x : 62
        y : 121
        width : 215
        height : 290
      },
      classes : [
        {
          Id : 58
          Class : 14
          Label : person
          Probability : 13,969749
          Classes : 20
        }, 
      ],
      predictions : [
        
      ]
    }, 
  ]
}

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
LABELS='labels.txt'
MODEL_LOCATION='graph_tinyyolov2.tflite'
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
tinyyolov2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:02.395698785 19446 0x555ab55d6770 LOG               tinyyolov2 gsttinyyolov2.c:286:gst_tinyyolov2_preprocess:<net> Preprocess
0:00:02.515331764 19446 0x555ab55d6770 LOG               tinyyolov2 gsttinyyolov2.c:325:gst_tinyyolov2_postprocess_old:<net> Postprocess
0:00:02.515377038 19446 0x555ab55d6770 LOG               tinyyolov2 gstinferencedebug.c:93:gst_inference_print_boxes:<net> Box: [class:14, x:97,778279, y:54,509112, width:229,643766, height:367,855935, prob:10,819336]
0:00:02.515401986 19446 0x555ab55d6770 LOG               tinyyolov2 gsttinyyolov2.c:359:gst_tinyyolov2_postprocess_new:<net> Postprocess Meta
0:00:02.515411728 19446 0x555ab55d6770 LOG               tinyyolov2 gsttinyyolov2.c:366:gst_tinyyolov2_postprocess_new:<net> Number of predictions: 1
0:00:02.515541193 19446 0x555ab55d6770 LOG               tinyyolov2 gstinferencedebug.c:111:gst_inference_print_predictions: 
{
  id : 24,
  enabled : True,
  bbox : {
    x : 0
    y : 0
    width : 416
    height : 416
  },
  classes : [
    
  ],
  predictions : [
    {
      id : 25,
      enabled : True,
      bbox : {
        x : 97
        y : 54
        width : 229
        height : 367
      },
      classes : [
        {
          Id : 18
          Class : 14
          Label : person
          Probability : 10,819336
          Classes : 20
        }, 
      ],
      predictions : [
        
      ]
    }, 
  ]
}

Visualization with detection overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_tinyyolov2.tflite'
LABELS='labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
tinyyolov2 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"  \
net.src_bypass ! inferenceoverlay style=2  font-scale=1 thickness=2 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output


TinyYolov3

Image file

  • Get the graph used on this example from RidgeRun Store
  • You will need an image file from one of TinyYOLO classes
  • Pipeline
IMAGE_FILE='cat.jpg'
LABELS='labels.txt'
MODEL_LOCATION='graph_tinyyolov3.tflite'
GST_DEBUG=tinyyolov3:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE start-index=0 stop-index=0 loop=true  ! jpegparse ! jpegdec ! videoconvert ! videoscale ! videorate ! queue ! net.sink_model \
tinyyolov3 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:06.401015400 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:06.817243785 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:06.817315935 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-55,170727, y:25,507316, width:396,182867, height:423,241143, prob:14,526398]
0:00:06.817426814 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:07.236310555 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:07.236379100 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-55,170727, y:25,507316, width:396,182867, height:423,241143, prob:14,526398]
0:00:07.236486242 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:07.659870194 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:07.659942388 12340      0x1317cf0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-55,170727, y:25,507316, width:396,182867, height:423,241143, prob:14,526398]

Video file

  • Get the graph used on this example from RidgeRun Store
  • You will need a video file from one of TinyYOLO classes
  • Pipeline
VIDEO_FILE='cat.mp4'
LABELS='labels.txt'
MODEL_LOCATION='graph_tinyyolov3.tflite'
GST_DEBUG=tinyyolov3:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! decodebin ! videoconvert ! videoscale ! queue ! net.sink_model \
tinyyolov3 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:08.545063684 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:08.955522899 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:08.955600820 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-36,012765, y:-37,118160, width:426,351621, height:480,353663, prob:14,378592]
0:00:08.955824676 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:09.364908234 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:09.364970901 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-36,490694, y:-38,108817, width:427,474399, height:482,318385, prob:14,257683]
0:00:09.365090340 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:09.775848590 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:09.775932404 12504       0xce4400 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:7, x:-35,991940, y:-37,482425, width:426,533537, height:480,917142, prob:14,313076]

Camera stream

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
LABELS='labels.txt'
MODEL_LOCATION='graph_tinyyolov3.tflite'
GST_DEBUG=tinyyolov3:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
tinyyolov3 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"
  • Output
0:00:06.823064776 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:07.242114002 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:07.242183276 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:14, x:116,796387, y:-31,424289, width:240,876587, height:536,305261, prob:11,859128]
0:00:07.242293677 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:07.660324555 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:07.660388215 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:14, x:113,453324, y:-27,681194, width:248,010337, height:528,964842, prob:11,603928]
0:00:07.660503502 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:479:gst_tinyyolov3_preprocess:<net> Preprocess
0:00:08.079154860 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:501:gst_tinyyolov3_postprocess:<net> Postprocess
0:00:08.079230404 12678       0xec24a0 LOG               tinyyolov3 gsttinyyolov3.c:384:print_top_predictions:<net> Box: [class:14, x:113,736444, y:-33,747251, width:246,987389, height:541,188374, prob:11,888664]

Visualization with detection overlay

  • Get the graph used on this example from RidgeRun Store
  • You will need a v4l2 compatible camera
  • Pipeline
CAMERA='/dev/video0'
MODEL_LOCATION='graph_tinyyolov3.tflite'
LABELS='labels.txt'
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! videoconvert ! tee name=t \
t. ! videoscale ! queue ! net.sink_model \
t. ! queue ! net.sink_bypass \
tinyyolov3 name=net model-location=$MODEL_LOCATION backend=tflite labels="$(cat $LABELS)"  \
net.src_bypass ! inferenceoverlay style=2  font-scale=1 thickness=2 ! videoconvert ! xvimagesink sync=false
  • Output
Classification overlay example output



Previous: Example pipelines Index Next: Example pipelines/NANO