Bounding box parser for retinanet

Please provide complete information as applicable to your setup.
• Hardware Platform: Jetson Xavier NX
• DeepStream Version: 6.0
• Language: Python

I trained a Retinanet model with TAO toolkit using custom dataset, and would like to port it on Jetson Xavier and run video analytics with Deepstream. To implement that, I need to first compile the bounding box parser. I was referring to this webpage, and used the code nvdsparsebbox_retinanet.cpp under this repository.

To build the custom parser and generate, I used the command

mkdir build && cd build 
cmake -DCMAKE_CUDA_FLAGS="--expt-extended-lambda -std=c++11" .. 

as in the technical blog.

It shows an error about missing file nvdsparsebbox_odtk.cpp, which I did not find under the repository.

UDA_FLAGS="--expt-extended-lambda -std=c++11" .. 
-- Configuring done
CMake Error at /usr/local/share/cmake-3.19/Modules/FindCUDA.cmake:1842 (add_library):
  Cannot find source file:


  Tried extensions .c .C .c++ .cc .cpp .cxx .cu .m .M .mm .h .hh .h++ .hm
  .hpp .hxx .in .txx .f .F .for .f77 .f90 .f95 .f03 .ispc
Call Stack (most recent call first):
  CMakeLists.txt:24 (cuda_add_library)

CMake Error at /usr/local/share/cmake-3.19/Modules/FindCUDA.cmake:1842 (add_library):
  No SOURCES given to target: nvdsparsebbox_odtk
Call Stack (most recent call first):
  CMakeLists.txt:24 (cuda_add_library)

CMake Generate step failed.  Build files cannot be regenerated correctly.

Is this repository up to date and supposed to work with DeepStream 6.0? Any suggestion is appreciated.

Hi @zzww ,
This project GitHub - NVIDIA-AI-IOT/deepstream_tao_apps: Sample apps to demonstrate how to deploy models trained with TAO on DeepStream supports TAO Retinanet with its parser, could you refer to this project?


Thanks for your reply. I will refer to the project and test it.

There is no update from you for a period, assuming this is not an issue anymore.
Hence we are closing this topic. If need further support, please open a new one.

Please let us know if you have more udpate, thanks.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.