Jump to content

GstInference/Example pipelines/TX2: Difference between revisions

GstInference/Example pipelines/TX2 (view source)

Revision as of 12:24, 21 May 2024

1,216 bytes added ,  21 May
m
no edit summary
mNo edit summary
mNo edit summary
 
Line 31: Line 31:
INPUT_LAYER='input'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE ! jpegparse ! nvjpegdec ! 'video/x-raw' ! nvvidconv ! 'video/x-raw(memory:NVMM),format=NV12' ! nvvidconv ! queue ! net.sink_model \
multifilesrc location=$IMAGE_FILE ! jpegparse ! nvjpegdec ! 'video/x-raw' ! nvvidconv ! 'video/x-raw(memory:NVMM),format=NV12' ! nvvidconv ! queue ! net.sink_model \
Line 62: Line 64:
INPUT_LAYER='input'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! qtdemux name=demux ! h264parse ! omxh264dec ! nvvidconv ! queue ! net.sink_model \
filesrc location=$VIDEO_FILE ! qtdemux name=demux ! h264parse ! omxh264dec ! nvvidconv ! queue ! net.sink_model \
Line 84: Line 88:
=== Inceptionv4 inference on camera stream using TensorFlow ===
=== Inceptionv4 inference on camera stream using TensorFlow ===
* Get the graph used on this example from [https://shop.ridgerun.com/products/inceptionv4-for-tensorflow this link]
* Get the graph used on this example from [https://shop.ridgerun.com/products/inceptionv4-for-tensorflow this link]
* You will need a camera compatible with Nvidia Libargus API or V4l2.
* You will need a camera compatible with NVIDIA Libargus API or V4l2.


====Nvidia Camera====
====NVIDIA Camera====


* Pipeline
* Pipeline
Line 94: Line 98:
INPUT_LAYER='input'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
nvcamerasrc sensor-id=$SENSOR_ID ! nvvidconv ! queue ! net.sink_model \
nvcamerasrc sensor-id=$SENSOR_ID ! nvvidconv ! queue ! net.sink_model \
Line 107: Line 113:
INPUT_LAYER='input'
INPUT_LAYER='input'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
Line 132: Line 140:
===Inceptionv4 visualization with classification overlay Tensorflow===
===Inceptionv4 visualization with classification overlay Tensorflow===
* Get the graph used on this example from [https://shop.ridgerun.com/products/inceptionv4-for-tensorflow this link]
* Get the graph used on this example from [https://shop.ridgerun.com/products/inceptionv4-for-tensorflow this link]
* You will need a camera compatible with Nvidia Libargus API or V4l2.
* You will need a camera compatible with NVIDIA Libargus API or V4l2.
====Nvidia Camera====
====NVIDIA Camera====
* Pipeline
* Pipeline
<syntaxhighlight lang=bash>
<syntaxhighlight lang=bash>
Line 141: Line 149:
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
LABELS='imagenet_labels.txt'
LABELS='imagenet_labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
gst-launch-1.0 \
gst-launch-1.0 \
nvcamerasrc sensor-id=$SENSOR_ID ! 'video/x-raw(memory:NVMM)' ! tee name=t \
nvcamerasrc sensor-id=$SENSOR_ID ! 'video/x-raw(memory:NVMM)' ! tee name=t \
Line 157: Line 167:
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
OUTPUT_LAYER='InceptionV4/Logits/Predictions'
LABELS='imagenet_labels.txt'
LABELS='imagenet_labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
gst-launch-1.0 \
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! tee name=t \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! tee name=t \
Line 164: Line 176:
net.src_bypass ! videoconvert ! classificationoverlay labels="$(cat $LABELS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
net.src_bypass ! videoconvert ! classificationoverlay labels="$(cat $LABELS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
</syntaxhighlight>
</syntaxhighlight>


* Output
* Output
Line 178: Line 189:
INPUT_LAYER='input/Placeholder'
INPUT_LAYER='input/Placeholder'
OUTPUT_LAYER='add_8'
OUTPUT_LAYER='add_8'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE ! jpegparse ! nvjpegdec ! 'video/x-raw' ! nvvidconv ! 'video/x-raw(memory:NVMM),format=NV12' ! nvvidconv ! queue ! net.sink_model \
multifilesrc location=$IMAGE_FILE ! jpegparse ! nvjpegdec ! 'video/x-raw' ! nvvidconv ! 'video/x-raw(memory:NVMM),format=NV12' ! nvvidconv ! queue ! net.sink_model \
Line 196: Line 209:
0:00:07.662473455 30513      0x5accf0 LOG              tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:07.662473455 30513      0x5accf0 LOG              tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:07.662769998 30513      0x5accf0 LOG              tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:7, x:25.820670, y:11.977936, width:425.495203, height:450.224357, prob:15.204609]
0:00:07.662769998 30513      0x5accf0 LOG              tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:7, x:25.820670, y:11.977936, width:425.495203, height:450.224357, prob:15.204609]
</syntaxhighlight>
</syntaxhighlight>


Line 208: Line 220:
INPUT_LAYER='input/Placeholder'
INPUT_LAYER='input/Placeholder'
OUTPUT_LAYER='add_8'
OUTPUT_LAYER='add_8'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! qtdemux name=demux ! h264parse ! omxh264dec ! nvvidconv ! queue ! net.sink_model \
filesrc location=$VIDEO_FILE ! qtdemux name=demux ! h264parse ! omxh264dec ! nvvidconv ! queue ! net.sink_model \
Line 230: Line 244:
=== TinyYolov2 inference on camera stream using Tensorflow ===
=== TinyYolov2 inference on camera stream using Tensorflow ===
* Get the graph used on this example from [https://shop.ridgerun.com/products/tinyyolov2-for-tensorflow this link]
* Get the graph used on this example from [https://shop.ridgerun.com/products/tinyyolov2-for-tensorflow this link]
* You will need a camera compatible with Nvidia Libargus API or V4l2.
* You will need a camera compatible with NVIDIA Libargus API or V4l2.
====Nvidia Camera====
====NVIDIA Camera====
* Pipeline
* Pipeline
<syntaxhighlight lang=bash>
<syntaxhighlight lang=bash>
Line 238: Line 252:
INPUT_LAYER='input/Placeholder'
INPUT_LAYER='input/Placeholder'
OUTPUT_LAYER='add_8'
OUTPUT_LAYER='add_8'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
nvarguscamerasrc sensor-id=$SENSOR_ID ! nvvidconv ! 'video/x-raw,format=BGRx' ! queue ! net.sink_model \
nvarguscamerasrc sensor-id=$SENSOR_ID ! nvvidconv ! 'video/x-raw,format=BGRx' ! queue ! net.sink_model \
Line 250: Line 266:
INPUT_LAYER='input/Placeholder'
INPUT_LAYER='input/Placeholder'
OUTPUT_LAYER='add_8'
OUTPUT_LAYER='add_8'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
Line 272: Line 290:
=== TinyYolov2 visualization with detection overlay Tensorflow ===
=== TinyYolov2 visualization with detection overlay Tensorflow ===
* Get the graph used on this example from [https://shop.ridgerun.com/products/tinyyolov2-for-tensorflow this link]
* Get the graph used on this example from [https://shop.ridgerun.com/products/tinyyolov2-for-tensorflow this link]
* You will need a camera compatible with Nvidia Libargus API or V4l2.
* You will need a camera compatible with NVIDIA Libargus API or V4l2.
====Nvidia Camera ====
====NVIDIA Camera ====
* Pipeline
* Pipeline
<syntaxhighlight lang=bash>
<syntaxhighlight lang=bash>
Line 281: Line 299:
OUTPUT_LAYER='add_8'
OUTPUT_LAYER='add_8'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=tinyyolov2:6 \
GST_DEBUG=tinyyolov2:6 \
gst-launch-1.0 \
gst-launch-1.0 \
Line 298: Line 318:
OUTPUT_LAYER='add_8'
OUTPUT_LAYER='add_8'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
gst-launch-1.0 \
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! tee name=t \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! tee name=t \
Line 305: Line 327:
net.src_bypass ! videoconvert ! detectionoverlay labels="$(cat $LABELS)" font-scale=1 thickness=2 ! videoconvert ! xvimagesink sync=false
net.src_bypass ! videoconvert ! detectionoverlay labels="$(cat $LABELS)" font-scale=1 thickness=2 ! videoconvert ! xvimagesink sync=false
</syntaxhighlight>
</syntaxhighlight>


* Output
* Output
Line 312: Line 333:
=== FaceNet visualization with embedding overlay Tensorflow ===
=== FaceNet visualization with embedding overlay Tensorflow ===
* Get the graph used on this example from [https://shop.ridgerun.com/products/facenetv1-for-tensorflow this link]
* Get the graph used on this example from [https://shop.ridgerun.com/products/facenetv1-for-tensorflow this link]
* You will need a camera compatible with Nvidia Libargus API or V4l2.
* You will need a camera compatible with NVIDIA Libargus API or V4l2.
* LABELS and EMBEDDINGS files are in $PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings.
* LABELS and EMBEDDINGS files are in $PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings.


====Nvidia Camera ====
====NVIDIA Camera ====
* Pipeline
* Pipeline
<syntaxhighlight lang=bash>
<syntaxhighlight lang=bash>
Line 324: Line 345:
LABELS='$PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings/labels.txt'
LABELS='$PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings/labels.txt'
EMBEDDINGS='$PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings/embeddings.txt'
EMBEDDINGS='$PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings/embeddings.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
gst-launch-1.0 \
gst-launch-1.0 \
nvcamerasrc sensor-id=$SENSOR_ID ! 'video/x-raw(memory:NVMM),width=(int)1280,height=(int)720' ! nvvidconv ! 'video/x-raw,format=BGRx,width=(int)1280,height=(int)720' ! videoconvert ! tee name=t \
nvcamerasrc sensor-id=$SENSOR_ID ! 'video/x-raw(memory:NVMM),width=(int)1280,height=(int)720' ! nvvidconv ! 'video/x-raw,format=BGRx,width=(int)1280,height=(int)720' ! videoconvert ! tee name=t \
Line 341: Line 364:
LABELS='$PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings/labels.txt'
LABELS='$PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings/labels.txt'
EMBEDDINGS='$PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings/embeddings.txt'
EMBEDDINGS='$PATH_TO_GST_INFERENCE_ROOT_DIR/tests/examples/embedding/embeddings/embeddings.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
gst-launch-1.0 \
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! tee name=t \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! tee name=t \
Line 362: Line 387:
MODEL_LOCATION='graph_inceptionv4.tflite'
MODEL_LOCATION='graph_inceptionv4.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE ! jpegparse ! nvjpegdec ! 'video/x-raw' ! nvvidconv ! 'video/x-raw(memory:NVMM),format=NV12' ! nvvidconv ! queue ! net.sink_model \
multifilesrc location=$IMAGE_FILE ! jpegparse ! nvjpegdec ! 'video/x-raw' ! nvvidconv ! 'video/x-raw(memory:NVMM),format=NV12' ! nvvidconv ! queue ! net.sink_model \
Line 380: Line 407:
0:02:22.678740356 30355      0x5accf0 LOG              inceptionv4 gstinceptionv4.c:232:gst_inceptionv4_postprocess:<net> Postprocess
0:02:22.678740356 30355      0x5accf0 LOG              inceptionv4 gstinceptionv4.c:232:gst_inceptionv4_postprocess:<net> Postprocess
0:02:22.678892356 30355      0x5accf0 LOG              inceptionv4 gstinceptionv4.c:253:gst_inceptionv4_postprocess:<net> Highest probability is label 282 : (0.627314)
0:02:22.678892356 30355      0x5accf0 LOG              inceptionv4 gstinceptionv4.c:253:gst_inceptionv4_postprocess:<net> Highest probability is label 282 : (0.627314)
</syntaxhighlight>
</syntaxhighlight>


Line 392: Line 418:
MODEL_LOCATION='graph_inceptionv4.tflite'
MODEL_LOCATION='graph_inceptionv4.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! qtdemux name=demux ! h264parse ! omxh264dec ! nvvidconv ! queue ! net.sink_model \
filesrc location=$VIDEO_FILE ! qtdemux name=demux ! h264parse ! omxh264dec ! nvvidconv ! queue ! net.sink_model \
Line 414: Line 442:
=== Inceptionv4 inference on camera stream using TensorFlow-Lite ===
=== Inceptionv4 inference on camera stream using TensorFlow-Lite ===
* Get the graph used on this example from [https://shop.ridgerun.com/products/inceptionv4-for-tensorflow-lite this link]
* Get the graph used on this example from [https://shop.ridgerun.com/products/inceptionv4-for-tensorflow-lite this link]
* You will need a camera compatible with Nvidia Libargus API or V4l2.
* You will need a camera compatible with NVIDIA Libargus API or V4l2.


====Nvidia Camera====
====NVIDIA Camera====


* Pipeline
* Pipeline
Line 423: Line 451:
MODEL_LOCATION='graph_inceptionv4.tflite'
MODEL_LOCATION='graph_inceptionv4.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
nvcamerasrc sensor-id=$SENSOR_ID ! nvvidconv ! queue ! net.sink_model \
nvcamerasrc sensor-id=$SENSOR_ID ! nvvidconv ! queue ! net.sink_model \
Line 435: Line 465:
MODEL_LOCATION='graph_inceptionv4.tflite'
MODEL_LOCATION='graph_inceptionv4.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
GST_DEBUG=inceptionv4:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
Line 460: Line 492:
===Inceptionv4 visualization with classification overlay TensorFlow-Lite===
===Inceptionv4 visualization with classification overlay TensorFlow-Lite===
* Get the graph used on this example from [https://shop.ridgerun.com/products/inceptionv4-for-tensorflow-lite this link]
* Get the graph used on this example from [https://shop.ridgerun.com/products/inceptionv4-for-tensorflow-lite this link]
* You will need a camera compatible with Nvidia Libargus API or V4l2.
* You will need a camera compatible with NVIDIA Libargus API or V4l2.
====Nvidia Camera====
====NVIDIA Camera====
* Pipeline
* Pipeline
<syntaxhighlight lang=bash>
<syntaxhighlight lang=bash>
Line 467: Line 499:
MODEL_LOCATION='graph_inceptionv4.tflite'
MODEL_LOCATION='graph_inceptionv4.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
gst-launch-1.0 \
gst-launch-1.0 \
nvcamerasrc sensor-id=$SENSOR_ID ! 'video/x-raw(memory:NVMM)' ! tee name=t \
nvcamerasrc sensor-id=$SENSOR_ID ! 'video/x-raw(memory:NVMM)' ! tee name=t \
Line 481: Line 515:
MODEL_LOCATION='graph_inceptionv4.tflite'
MODEL_LOCATION='graph_inceptionv4.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
gst-launch-1.0 \
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! tee name=t \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! tee name=t \
Line 488: Line 524:
net.src_bypass ! videoconvert ! classificationoverlay labels="$(cat $LABELS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
net.src_bypass ! videoconvert ! classificationoverlay labels="$(cat $LABELS)" font-scale=4 thickness=4 ! videoconvert ! xvimagesink sync=false
</syntaxhighlight>
</syntaxhighlight>


* Output
* Output
Line 501: Line 536:
MODEL_LOCATION='graph_tinyyolov2.tflite'
MODEL_LOCATION='graph_tinyyolov2.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
multifilesrc location=$IMAGE_FILE ! jpegparse ! nvjpegdec ! 'video/x-raw' ! nvvidconv ! 'video/x-raw(memory:NVMM),format=NV12' ! nvvidconv ! queue ! net.sink_model \
multifilesrc location=$IMAGE_FILE ! jpegparse ! nvjpegdec ! 'video/x-raw' ! nvvidconv ! 'video/x-raw(memory:NVMM),format=NV12' ! nvvidconv ! queue ! net.sink_model \
Line 519: Line 556:
0:00:07.662473455 30513      0x5accf0 LOG              tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:07.662473455 30513      0x5accf0 LOG              tinyyolov2 gsttinyyolov2.c:501:gst_tinyyolov2_postprocess:<net> Postprocess
0:00:07.662769998 30513      0x5accf0 LOG              tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:7, x:25.820670, y:11.977936, width:425.495203, height:450.224357, prob:15.204609]
0:00:07.662769998 30513      0x5accf0 LOG              tinyyolov2 gsttinyyolov2.c:384:print_top_predictions:<net> Box: [class:7, x:25.820670, y:11.977936, width:425.495203, height:450.224357, prob:15.204609]
</syntaxhighlight>
</syntaxhighlight>


Line 530: Line 566:
MODEL_LOCATION='graph_tinyyolov2.tflite'
MODEL_LOCATION='graph_tinyyolov2.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
filesrc location=$VIDEO_FILE ! qtdemux name=demux ! h264parse ! omxh264dec ! nvvidconv ! queue ! net.sink_model \
filesrc location=$VIDEO_FILE ! qtdemux name=demux ! h264parse ! omxh264dec ! nvvidconv ! queue ! net.sink_model \
Line 552: Line 590:
=== TinyYolov2 inference on camera stream using TensorFlow-Lite ===
=== TinyYolov2 inference on camera stream using TensorFlow-Lite ===
* Get the graph used on this example from [https://shop.ridgerun.com/products/tinyyolov2-for-tensorflow-lite this link]
* Get the graph used on this example from [https://shop.ridgerun.com/products/tinyyolov2-for-tensorflow-lite this link]
* You will need a camera compatible with Nvidia Libargus API or V4l2.
* You will need a camera compatible with NVIDIA Libargus API or V4l2.
====Nvidia Camera====
====NVIDIA Camera====
* Pipeline
* Pipeline
<syntaxhighlight lang=bash>
<syntaxhighlight lang=bash>
Line 559: Line 597:
MODEL_LOCATION='graph_tinyyolov2.tflite'
MODEL_LOCATION='graph_tinyyolov2.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
nvarguscamerasrc sensor-id=$SENSOR_ID ! nvvidconv ! 'video/x-raw,format=BGRx' ! queue ! net.sink_model \
nvarguscamerasrc sensor-id=$SENSOR_ID ! nvvidconv ! 'video/x-raw,format=BGRx' ! queue ! net.sink_model \
Line 570: Line 610:
MODEL_LOCATION='graph_tinyyolov2.tflite'
MODEL_LOCATION='graph_tinyyolov2.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
GST_DEBUG=tinyyolov2:6 gst-launch-1.0 \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
v4l2src device=$CAMERA ! videoconvert ! videoscale ! queue ! net.sink_model \
Line 592: Line 634:
=== TinyYolov2 visualization with detection overlay TensorFlow-Lite ===
=== TinyYolov2 visualization with detection overlay TensorFlow-Lite ===
* Get the graph used on this example from [https://shop.ridgerun.com/products/tinyyolov2-for-tensorflow-lite this link]
* Get the graph used on this example from [https://shop.ridgerun.com/products/tinyyolov2-for-tensorflow-lite this link]
* You will need a camera compatible with Nvidia Libargus API or V4l2.
* You will need a camera compatible with NVIDIA Libargus API or V4l2.
====Nvidia Camera ====
====NVIDIA Camera ====
* Pipeline
* Pipeline
<syntaxhighlight lang=bash>
<syntaxhighlight lang=bash>
Line 599: Line 641:
MODEL_LOCATION='graph_tinyyolov2.tflite'
MODEL_LOCATION='graph_tinyyolov2.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
GST_DEBUG=tinyyolov2:6 \
GST_DEBUG=tinyyolov2:6 \
gst-launch-1.0 \
gst-launch-1.0 \
Line 614: Line 658:
MODEL_LOCATION='graph_tinyyolov2.tflite'
MODEL_LOCATION='graph_tinyyolov2.tflite'
LABELS='labels.txt'
LABELS='labels.txt'
</syntaxhighlight>
<syntaxhighlight lang=bash>
gst-launch-1.0 \
gst-launch-1.0 \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! tee name=t \
v4l2src device=$CAMERA ! "video/x-raw, width=1280, height=720" ! tee name=t \
Retrieved from "https://developer.ridgerun.com/wiki/index.php/Special:MobileDiff/54349"
Cookies help us deliver our services. By using our services, you agree to our use of cookies.