Research map / Label-efficient and robust learning / Recognition and long-tail
Recognition of Action Units in the Wild With Deep Nets and a New Global-Local Loss
C. Fabian Benitez-Quiroz, Yan Wang, Aleix M. Martinez
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add the global term to the loss (GL-loss) instead of local loss only (ablation) | GoogLeNet-based AU recognition net · EmotioNet training | number of training epochs: reduced the number of epochs by more than 25% | Improved |
| train a single AlexNet for all AUs instead of one network per AU | AlexNet softmax baseline · EmotioNet challenge | Final score | Hurt |
| reduce image size to 1/2 and 1/4 of original | GL-loss deep net · EmotioNet scale test | Final score | No clear effect |
Closest papers by meaning
- Deep Facial Action Unit Recognition From Partially Labeled Data (ICCV 2017 · similarity 0.88)
- Action Unit Detection With Region Adaptation, Multi-Labeling Learning and Optimal Temporal Fusing (CVPR 2017 · similarity 0.87)
- Deep Structure Inference Network for Facial Action Unit Recognition (ECCV 2018 · similarity 0.87)
- Multi-label Co-regularization for Semi-supervised Facial Action Unit Recognition (NeurIPS 2019 · similarity 0.86)
- Action Recognition With Spatial-Temporal Discriminative Filter Banks (ICCV 2019 · similarity 0.86)
- Action Unit Memory Network for Weakly Supervised Temporal Action Localization (CVPR 2021 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
