search

researchmm / LightTrack

[CVPR21] LightTrack: Finding Lightweight Neural Network for Object Tracking via One-Shot Architecture Search
MIT License
396 stars 59 forks source link

About inference speed #6

Closed Youjiangbaba closed 3 years ago

Youjiangbaba commented 3 years ago

Hello ,your work is very wonderful and testing in my video achieved a good performance. One frame inference time is about 13ms and ocean is about 15ms in 2080Ti. How to raise the fps? Thanks for your open-source codes and methods and look forward your prompt reply.

MasterBin-IIAU commented 3 years ago

Hi, thanks for your appreciation to our work :)

First, about the speed on the Desktop GPUs (such as RTX 2080 Ti, RTX TITAN), the speed advantage of LightTrack to Ocean is not quite obvious, because the standard conv (especially 3x3 conv) is highly optimized and runs very fast on these platforms. However, the core operations of the LightTrack are 3x3 depth-wise conv and 1x1 conv. These operations run much slower than the standard 3x3 conv, although they have much less FLOPs.

Second, about how to raise the FPS, we suggest using tools like ONNX or TensorRT to speed-up the inference.

Youjiangbaba commented 3 years ago

Ok, I got it , thanks for your reply. Wish you a happy life!