mlsanityData to 7 Oct 2026

Multi-Level Adaptive Region of Interest and Graph Learning for Facial Action Unit Recognition

arXiv 2102.12154 ↗First posted 24 Feb 2021cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

In facial action unit (AU) recognition tasks, regional feature learning and AU relation modeling are two effective aspects which are worth exploring. However, the limited representation capacity of regional features makes it difficult for relation models to embed AU relationship knowledge. In this paper, we propose a novel multi-level adaptive ROI and graph learning (MARGL) framework to tackle this problem. Specifically, an adaptive ROI learning module is designed to automatically adjust the location and size of the predefined AU regions. Meanwhile, besides relationship between AUs, there exists strong relevance between regional features across multiple levels of the backbone network as level-wise features focus on different aspects of representation. In order to incorporate the intra-level AU relation and inter-level AU regional relevance simultaneously, a multi-level AU relation graph is constructed and graph convolution is performed to further enhance AU regional features of each level. Experiments on BP4D and DISFA demonstrate the proposed MARGL significantly outperforms the previous state-of-the-art methods.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 12 of its 20 references that are in the map. First posted on arXiv 24 Feb 2021.

Builds on

12 of 20 references.

References · 12

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  3. 2015Spatial Transformer Networkscited by 1,423
  4. 2018Deep Adaptive Attention for Joint Facial Action Unit Detection and Face Alignmentcited by 55
  5. 2019Semantic Relationships Guided Representation Learning for Facial Action Unit Recognitioncited by 53
  6. 2018Facial Action Unit Detection Using Attention and Relation Learningcited by 46
  7. 2018Deep Structure Inference Network for Facial Action Unit Recognitioncited by 38
  8. 2017Action Unit Detection with Region Adaptation, Multi-labeling Learning and Optimal Temporal Fusingcited by 28
  9. 2017EAC-Net: A Region-based Deep Enhancing and Cropping Approach for Facial Action Unit Detectioncited by 26
  10. 2019Multi-label Co-regularization for Semi-supervised Facial Action Unit Recognitioncited by 26
  11. 2019Relation Modeling with Graph Convolutional Networks for Facial Action Unit Detectioncited by 25
  12. 2017Deep Structured Learning for Facial Action Unit Intensity Estimationcited by 16

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 1; 2023: 1.

3 papers.

  1. 2022Weakly Supervised Regional and Temporal Learning for Facial Action Unit Recognitioncited by 3
  2. 2021Self-Supervised Regional and Temporal Auxiliary Tasks for Facial Action Unit Recognitioncited by 1
  3. 2023Boosting Facial Action Unit Detection Through Jointly Learning Facial Landmark Detection and Domain Separation and Reconstructioncited by 0