Deployment of BEVDepth on Nvidia board with onnx conversion
zhoumu1990 opened this issue · 1 comments
zhoumu1990 commented
Dear blogger, have you ever tried or planed to deloy BEVDepth on Orin board?
Two problems I've confronted in this process are :
- when converting to onnx, the process stopped with function VoxelPoolingInference();
- is multi-threaded CUDA inference supported for Orin board?
Look forward to your reply. Thanks:)