search

YonghaoHe / LFFD-A-Light-and-Fast-Face-Detector-for-Edge-Devices

A light and fast one class detection framework for edge devices. We provide face detector, head detector, pedestrian detector, vehicle detector......
MIT License
1.32k stars 329 forks source link

Lighter backbone for face detection #23

Closed lufi1 closed 3 years ago

lufi1 commented 5 years ago

Hi,

First of all, thank you for the project, it is super good and useful.

In the face detection repo, you say that you'll try :

much lighter backbone. MobileNet and ShuffleNet series are classic choices, maybe they are

not the best. We will explore something new and touch the limit of the speed.

Do you have any ideas about when this will be ready or how I could do it myself.

Many thanks

YonghaoHe commented 5 years ago

@lufi1 Reducing the size of feature maps as early as possible is the easiest way to decrease the inference latency. As we know, reducing the size of feature maps will harm the detection of small objects. So our direction is to reducing the size while keeping the accuracy as high as possible. Of cause, you can try MobileNet or other light backbones. The code is easy to modify and play with if you are familiar with MXNet.