Multi-modal Multi-label Facial Action Unit Detection with Transformer (ABAW2022 submission)

(Submission to the Affective Behavior Analysis in-the-wild (ABAW3) 2022 competition)

This repository presents a Multi-modal Transformer based model for Multi-label Facial Action Unit Detection

Pre-trained Models

The pretrained model is available at： Model weight with extract code: 7x18

torch 1.6.0, torchaudio 0.6, tqdm, Numpy, OpenCV 4.2.0 lmdb einops

part of code is from

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
112_align		112_align
dataloader		dataloader
metrics		metrics
models		models
postprocess		postprocess
LICENSE		LICENSE
README.md		README.md
opts.py		opts.py
test_aff2.py		test_aff2.py
train.py		train.py
utils.py		utils.py