NVIDIA VPI GStreamer Plug-in - Performance - Jetson Nano
| GstVPI |
|---|
|
| GstVPI Basics |
| Getting Started |
| Examples |
| Performance |
| Contact Us |
 | JetPack version 4.4.1 and VPI version 0.4.4 is used for the performance measurements. |
NVIDIA VPI GStreamer Plug-in Performance Measurements: Source - RasberryPi HQ Camera (IMX477 Sensor)
Algorithm : Gaussian Filter
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a Gaussian filter given by kernel size of 3 in the X direction and 3 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpigaussianfilter size-x=3 size-y=3 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpigaussianfilter size-x=3 size-y=3 backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 16.1786 % | 19.5862 % @ 99.8966 MHz | 30.044 |
| CPU | 27.8214 % | 31.2802 % @ 256.856 MHz | 30.041 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a Gaussian filter given by kernel size of 5 in the X direction and 5 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpigaussianfilter size-x=5 size-y=5 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpigaussianfilter size-x=5 size-y=5 backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 15.7857 % | 18.0805 % @ 102.552 MHz | 30.039 |
| CPU | 28.2069 % | 31.2802 % @ 256.856 MHz | 30.045 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a Gaussian filter given by kernel size of 7 in the X direction and 7 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpigaussianfilter size-x=7 size-y=7 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpigaussianfilter size-x=7 size-y=7 backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 15.5862 % | 20 % @ 105.207 MHz | 30.022 |
| CPU | 29.25 % | 31.2802 % @ 256.856 MHz | 30.052 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a Gaussian filter given by kernel size of 11 in the X direction and 11 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpigaussianfilter size-x=11 size-y=11 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpigaussianfilter size-x=11 size-y=11 backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 16.3571 % | 22.1241 % @ 117.421 MHz | 30.027 |
| CPU | 29.3103 % | 31.2802 % @ 256.856 MHz | 30.039 |
Algorithm : Undistort
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiundistort k1=0.05 k2=0.01 k3=0 k4=0 model=fisheye ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiundistort k1=0.05 k2=0.01 k3=0 k4=0 model=fisheye backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 19.069 % | 25.6897 % @ 175.408 MHz | 30.075 |
| CPU | 83.5926 % | 31.2802 % @ 256.856 MHz | 11.077 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiundistort k1=0.52 k2=-0.02 p1=0 p2=0 k3=-0.001 k4=0.8 k5=0.05 k6=-0.008 model=polynomial ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiundistort k1=0.52 k2=-0.02 p1=0 p2=0 k3=-0.001 k4=0.8 k5=0.05 k6=-0.008 model=polynomial backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 19.2143 % | 28.3645 % @ 214.177 MHz | 30.094 |
| CPU | 83.7037 % | 31.2802 % @ 256.856 MHz | 11.075 |
Algorithm : KLT Tracker
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1280x720
Backend : CUDA (default)
gst-launch-1.0 filesrc location=/opt/nvidia/vpi/samples/assets/dashcam.mp4 ! qtdemux name=demux demux.video_0 ! queue ! h264parse ! omxh264dec ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiklttracker name=klt boxes="<<613,332,23,23>,<669,329,30,29>,<790,376,41,22>>" ! perf print-arm-load=true ! vpidownload ! nvvidconv ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 filesrc location=/opt/nvidia/vpi/samples/assets/dashcam.mp4 ! qtdemux name=demux demux.video_0 ! queue ! h264parse ! omxh264dec ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiklttracker name=klt boxes="<<613,332,23,23>,<669,329,30,29>,<790,376,41,22>>" backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv ! ""video/x-raw(memory:NVMM),format=I420"" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 21.0952 % | 32.0106 % @ 257.204 MHz | 29.776 |
| CPU | 26.2857 % | 31.2802 % @ 256.856 MHz | 29.99 |
Algorithm : Harris Detector
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiharrisdetector gradient-size=3 block-size=3 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiharrisdetector gradient-size=3 block-size=3 backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 17.3103 % | 32.4612 % @ 242.922 MHz | 29.246 |
| CPU | 76.8621 % | 31.2802 % @ 256.856 MHz | 21.458 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiharrisdetector gradient-size=5 block-size=5 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiharrisdetector gradient-size=5 block-size=5 backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 19.3793 % | 34.8659 % @ 272.946 MHz | 29.986 |
| CPU | 80.2759 % | 31.2802 % @ 256.856 MHz | 17.813 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiharrisdetector gradient-size=7 block-size=7 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiharrisdetector gradient-size=7 block-size=7 backend = 1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 17.069 % | 37.3241 % @ 297.497 MHz | 28.99 |
| CPU | 83.2857 % | 31.2802 % @ 256.856 MHz | 14.457 |
Algorithm : Convolution
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a convolution filter given by kernel size of 3 in the X direction and 3 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiconvolution kernel="<<0.1111,0.1111,0.1111>,<0.1111,0.1111,0.1111>,<0.1111,0.1111,0.1111>>" ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiconvolution kernel="<<0.1111,0.1111,0.1111>,<0.1111,0.1111,0.1111>,<0.1111,0.1111,0.1111>>" backend = 1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 16.6429 % | 4.89383 % @ 105.161 MHz | 30.034 |
| CPU | 31.3333 % | 31.2802 % @ 256.856 MHz | 30.048 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a convolution filter given by kernel size of 5 in the X direction and 5 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiconvolution kernel="<<0.04,0.04,0.04,0.04,0.04>,<0.04,0.04,0.04,0.04,0.04>,<0.04,0.04,0.04,0.04,0.04>,<0.04,0.04,0.04,0.04,0.04>,<0.04,0.04,0.04,0.04,0.04>>" ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiconvolution kernel="<<0.04,0.04,0.04,0.04,0.04>,<0.04,0.04,0.04,0.04,0.04>,<0.04,0.04,0.04,0.04,0.04>,<0.04,0.04,0.04,0.04,0.04>,<0.04,0.04,0.04,0.04,0.04>>" backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 15.2857 % | 5.80854 % @ 105.375 MHz | 30.045 |
| CPU | 37.2069 % | 31.2802 % @ 256.856 MHz | 30.039 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a convolution filter given by kernel size of 7 in the X direction and 7 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiconvolution kernel="<<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>>" ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiconvolution kernel="<<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>,<0.0204,0.0204,0.0204,0.0204,0.0204,0.0204,0.0204>>" backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 16.2857 % | 7.08546 % @ 109.487 MHz | 30.052 |
| CPU | 44.1786 % | 31.2802 % @ 256.856 MHz | 30.045 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a convolution filter given by kernel size of 11 in the X direction and 11 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiconvolution kernel="<<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>>" ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiconvolution kernel="<<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>,<0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083,0.0083>>" backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 15.2414 % | 8.07749 % @ 111.986 MHz | 30.035 |
| CPU | 74.8276 % | 31.2802 % @ 256.856 MHz | 24.453 |
Algorithm : Separable Convolution
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a separable convolution filter given by kernel size of 3 in the X direction and 3 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiseparableconvolution kernel-row="<-1.0,0.0,1.0>" kernel-col="<1.0,2.0,1.0>" ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiseparableconvolution kernel-row="<-1.0,0.0,1.0>" kernel-col="<1.0,2.0,1.0>" backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 16.0357 % | 17.3563 % @ 99.8966 MHz | 30.039 |
| CPU | 31 % | 31.2802 % @ 256.856 MHz | 30.026 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a separable convolution filter given by kernel size of 5 in the X direction and 5 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiseparableconvolution kernel-row="<-2.0,-1.0,0.0,1.0,2.0>" kernel-col="<1.0,2.0,4.0,2.0,1.0>" ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiseparableconvolution kernel-row="<-2.0,-1.0,0.0,1.0,2.0>" kernel-col="<1.0,2.0,4.0,2.0,1.0>" backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 16.2143 % | 17.2931 % @ 99.2328 MHz | 30.036 |
| CPU | 32 % | 31.2802 % @ 256.856 MHz | 30.041 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a separable convolution filter given by kernel size of 7 in the X direction and 7 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiseparableconvolution kernel-row="<-4.0,-2.0,-1.0,0.0,1.0,2.0,4.0>" kernel-col="<1.0,2.0,4.0,6.0,4.0,2.0,1.0>" ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiseparableconvolution kernel-row="<-4.0,-2.0,-1.0,0.0,1.0,2.0,4.0>" kernel-col="<1.0,2.0,4.0,6.0,4.0,2.0,1.0>" backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 15.4286 % | 22.4741 % @ 108.858 MHz | 30.031 |
| CPU | 34.7857 % | 31.2802 % @ 256.856 MHz | 30.038 |
Platform : Jetson Nano developer kit
RasberryPi HQ Camera (IMX477 Sensor)
Resolution: 1920x1080
For a separable convolution filter given by kernel size of 11 in the X direction and 11 in the Y direction.
Backend : CUDA (default)
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiseparableconvolution kernel-row="<-16.0,-8.0,-4.0,-2.0,-1.0,0.0,1.0,2.0,4.0,8.0,16.0>" kernel-col="<1.0,2.0,4.0,6.0,12.0,24.0,12.0,6.0,4.0,2.0,1.0>" ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
Backend : CPU
gst-launch-1.0 nvarguscamerasrc ! nvvidconv ! 'video/x-raw' ! vpiupload ! vpiseparableconvolution kernel-row="<-16.0,-8.0,-4.0,-2.0,-1.0,0.0,1.0,2.0,4.0,8.0,16.0>" kernel-col="<1.0,2.0,4.0,6.0,12.0,24.0,12.0,6.0,4.0,2.0,1.0>" backend=1 ! perf print-arm-load=true ! vpidownload ! nvvidconv flip-method=2 ! "video/x-raw(memory:NVMM),format=I420" ! nvoverlaysink
| Backend | CPU % | GPU % | FPS |
|---|---|---|---|
| CUDA | 16.5517 % | 23.1801 % @ 109.927 MHz | 30.041 |
| CPU | 39.3214 % | 31.2802 % @ 256.856 MHz | 30.043 |