search

marcoslucianops / DeepStream-Yolo

NVIDIA DeepStream SDK 7.0 / 6.4 / 6.3 / 6.2 / 6.1.1 / 6.1 / 6.0.1 / 6.0 / 5.1 implementation for YOLO models
MIT License
1.46k stars 355 forks source link

Is batch_nms basically included in the TensorRT build? #247

Open YoungjaeDev opened 2 years ago

YoungjaeDev commented 2 years ago

Hello I can see the batch_nms layer at the end of the TensorRT build, is this basically included? What can I do if I want to exclude this?

YoungjaeDev commented 2 years ago

Oh, it's gone because I recently updated the code commits. nms basically working when cluster_mode=4 in the TensorRT model? I want to use the post-processing code that I made, but I want to use as little resources as possible when cluster_mode=4

marcoslucianops commented 2 years ago

I reverted the NMS to DeepStream default NMS due to accurary drop, so the Batched NMS isn't used anymore.