CarkusL

> Hello, can you please tell me how can I display the Precision and Recall values, I need them for my paper, thank you. You can use my branch https://github.com/CarkusL/SFA3D

I download latest.pth from https://github.com/tianweiy/CenterPoint/blob/master/configs/nusc/README.md I think it trained on v1.0-trainval

I uploaded the visualisation scripts TensorRT_Visualize.ipynb . You can check the ReadMe https://github.com/CarkusL/CenterPoint/tree/main/tensorrt/samples/centerpoint

I will add the postprocess implementation in TensorRT in the next few weeks.

> Is your pipeline complete? I mean including the postprocessing part? yes, please check the postprocess.cpp

I didn't try to export voxelnet to onnx. The ONNX and Tensorrt don't support the sparse convolution.

CUDA : 11.1 TensorRT : 7.2.2-1 GPU : RTX3090

> Hi @CarkusL, can you give us the inference time for batch_size =1 of your TensorRT implementation including also the preprocess and postprocess? [09/15/2021-10:38:05] [I] PreProcess Time: 13.3244 ms [09/15/2021-10:38:05]...

> > > Hi @CarkusL, can you give us the inference time for batch_size =1 of your TensorRT implementation including also the preprocess and postprocess? > > > > >...

no, it doesn't support training on kitti