search

YonghaoHe / LFD-A-Light-and-Fast-Detector

LFD is a big update upon LFFD. Generally, LFD is a multi-class object detector characterized by lightweight, low inference latency and superior precision. It is for real-world appilcations.
419 stars 82 forks source link

May you release a cpu-based inference version? #15

Closed on-your-way closed 3 years ago

on-your-way commented 3 years ago

As is known to all, it is better using CUDA for the training procedure. However, it is not practical for inference since many edge devices have not a GPU/CUDA hardway, such as my Surface. In your README file, it is your wish that making LFD popular as YOLO series. So, I suggest that you release a cpu-based (even arm-based) inference version for edge devices, just like the Libfacedetection.

YonghaoHe commented 3 years ago

@on-your-way you mean Intel CPU?

on-your-way commented 3 years ago

@on-your-way you mean Intel CPU?

yes

lucasjinreal commented 3 years ago

@on-your-way What do u suggest way to deploy CPU version? libtorch? tvm? openvino?

YonghaoHe commented 3 years ago

@on-your-way @jinfagang OPENVINO is the first choice for deployment on Intel CPU. You can implement yourself.

lucasjinreal commented 3 years ago

@YonghaoHe I am current trained some other dataset found that FP rate of LFD is a little high.

on-your-way commented 3 years ago

@on-your-way What do u suggest way to deploy CPU version? libtorch? tvm? openvino?

Thanks for suggesting. I am used to use the DNN module of OpenCV and have not tried others yet.

on-your-way commented 3 years ago

@YonghaoHe Thanks. I will try it.