Summary of Learning Contrastive Feature Representations For Facial Action Unit Detection, by Ziqiao Shang et al.
Learning Contrastive Feature Representations for Facial Action Unit Detection
by Ziqiao Shang, Bin Liu, Fengmao Lv, Fei Teng, Tianrui Li
First submitted to arxiv on: 9 Feb 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel contrastive learning framework for facial action unit (AU) detection. The framework combines self-supervised and supervised signals to learn discriminative features for accurate AU detection. To address class imbalance, the paper employs a negative sample re-weighting strategy. Additionally, it uses a sampling technique to mitigate the effects of noisy labels. Experimental results on five benchmark datasets show that the proposed approach outperforms state-of-the-art methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper helps us better detect facial actions, like smiling or frowning. It’s hard because some emotions look similar, and there are lots of fake or wrong labels mixed in. The researchers came up with a new way to learn from data that includes both correct and incorrect labels. This helps the model ignore the bad labels and focus on the good ones. They tested their method on many famous datasets and showed it works better than other methods. |
Keywords
* Artificial intelligence * Self supervised * Supervised
