Python Tracking on Jetson

Autonomous Machines Jetson & Embedded Systems Jetson AGX Xavier
1

Hi, I want use object tracking and detection in my python app, and have a problem with computing resources. When used detection only fps was near 100 on MobileNet inference, but i added dlib or mosse(pyhton) tracking fps became equal 1. What am I doing wrong on my Jetson Xavier?

2

Hi,

This may be related to the dlib implementation.
You can get some profiling data for your program first.

nvprof python3 [app.py]

Thanks.

3 
nvprof python3 my-detection.py 
jetson.inference.__init__.py
jetson.inference -- initializing Python 3.6 bindings...
jetson.inference -- registering module types...
jetson.inference -- done registering module types
jetson.inference -- done Python 3.6 binding initialization
jetson.utils.__init__.py
jetson.utils -- initializing Python 3.6 bindings...
jetson.utils -- registering module functions...
jetson.utils -- done registering module functions
jetson.utils -- registering module types...
jetson.utils -- done registering module types
jetson.utils -- done Python 3.6 binding initialization
jetson.inference -- PyTensorNet_New()
jetson.inference -- PyDetectNet_Init()
jetson.inference -- detectNet loading build-in network 'ssd-inception-v2'

detectNet -- loading detection network model from:
          -- model        networks/SSD-Inception-v2/ssd_inception_v2_coco.uff
          -- input_blob   'Input'
          -- output_blob  'NMS'
          -- output_count 'NMS_1'
          -- class_labels networks/SSD-Inception-v2/ssd_coco_labels.txt
          -- threshold    0.400000
          -- batch_size   1

[TRT]   TensorRT version 6.0.1
[TRT]   loading NVIDIA plugins...
[TRT]   Plugin Creator registration succeeded - GridAnchor_TRT
[TRT]   Plugin Creator registration succeeded - GridAnchorRect_TRT
[TRT]   Plugin Creator registration succeeded - NMS_TRT
[TRT]   Plugin Creator registration succeeded - Reorg_TRT
[TRT]   Plugin Creator registration succeeded - Region_TRT
[TRT]   Plugin Creator registration succeeded - Clip_TRT
[TRT]   Plugin Creator registration succeeded - LReLU_TRT
[TRT]   Plugin Creator registration succeeded - PriorBox_TRT
[TRT]   Plugin Creator registration succeeded - Normalize_TRT
[TRT]   Plugin Creator registration succeeded - RPROI_TRT
[TRT]   Plugin Creator registration succeeded - BatchedNMS_TRT
[TRT]   Could not register plugin creator:  FlattenConcat_TRT in namespace: 
[TRT]   completed loading NVIDIA plugins.
[TRT]   detected model format - UFF  (extension '.uff')
[TRT]   desired precision specified for GPU: FASTEST
[TRT]   requested fasted precision for device GPU without providing valid calibrator, disabling INT8
==9741== NVPROF is profiling process 9741, command: python3 my-detection.py
==9741== Warning: Insufficient privileges to start the profiling session. Use root privileges
==9741== Profiling application: python3 my-detection.py
==9741== Profiling result:
No kernels were profiled.
No API activities were profiled.
4

Try exec sudo and from root, result:
nvprof: command not found

5

Hi,

Please update to this command:

/usr/local/cuda-10.0/bin/nvprof python3 [app.py] -o data.nvvp

And share the data.nvvp with us.

Thanks.