MOS: A Low Latency and Lightweight Framework for Face Detection, Landmark Localization, and Head Pose Estimation
Introduction
This repo is the official implementation of “MOS: A Low Latency and Lightweight Framework for Face Detection, Landmark Localization, and Head Pose Estimation”. The paper has been accepted at BMVC2021.
This repo is an implementation of PyTorch. MOS is a low latency and lightweight architecture for face detection, facial landmark localization and head pose estimation.It aims to bridge the gap between research and industrial communities.
For more details, please refer to our report on Arxiv.
Updates
- 【2021/10/31】 We have released the training data (widerface with pose label). The pytorch inference code of MOS-S and MOS-M has been released!
- 【2021/10/22】 We have released our paper on Arxiv.