yukang

Hi, Thank you very much for your notification. I am currently on a vacation. I will try this next week. Thanks for your understanding. Regards, Yukang Chen

Hi, I think this might be hard. Because the spconv backbone receives sparse tensors, which is not supported yet by ONNX. I have converted the head of VoxelNeXt to ONNX....

Hi, Thanks for your interest in our work. The tracking part will be released latter. I am preparing the code. Thanks for your understanding. Regards, Yukang Chen

Hi Naren, You can try to directly remove these two sentences in the [pcdet](https://github.com/dvlab-research/VoxelNeXt/tree/master/pcdet)/[utils](https://github.com/dvlab-research/VoxelNeXt/tree/master/pcdet/utils)/spconv_utils.py ``` if float(spconv.__version__[2:]) >= 2.2: spconv.constants.SPCONV_USE_DIRECT_TABLE = False ``` Regards, Yukang Chen

Hi, What is your version of spconv? Regards, Yukang Chen

Please try a new spconv version. For example, pip install spconv==2.1.25

Hi, I think this is the problem of `kornia`. Would you please check its version? Regards, Yukang Chen

Hi, Have you ever installed OpenPCDet? `python3 setup.py develop` Sorry. I did not use docker. Regards, Yukang Chen

Hi, what is the output from python setup.py develop? In addition, it seems that the problem is still in the installation of openpcdet.

You can change the backbone name to be "VoxelResBackBone8xVoxelNeXtSPS" in the config file.