Research map / Label-efficient and robust learning / Domain adaptation and generalization
Generalizable Facial Expression Recognition
Yuhang Zhang, Xiuqi Zheng, Chenyi Liang, Jiani Hu, Weihong Deng
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 |
|---|---|---|---|
| learn sigmoid masks on fixed CLIP face features (ablation) | CAFE (ResNet-18) · Train RAF-DB, mean over four unseen test sets | mean accuracy: 46.67 -> 55.49 | Improved |
| add channel-separation module (per-class channel groups, no FC) (ablation) | CAFE · Train RAF-DB, mean over four unseen test sets | mean accuracy: 55.49 -> 56.80 | Improved |
| add channel-diverse loss (ablation) | CAFE · Train RAF-DB, mean over four unseen test sets | mean accuracy: 56.80 -> 57.17 | Improved |
| apply sigmoid to learned masks (ablation) | CAFE · Train RAF-DB, mean over five test sets | mean accuracy: 58.05 -> 63.48 | Improved |
| apply sigmoid mask: effect on source domain RAF-DB (ablation) | CAFE · Train RAF-DB, test RAF-DB (source) | accuracy: 89.24 -> 88.72 | Mixed |
| use CAFE mask design instead of fine-tuning FC on concatenated CLIP features (ablation) | CLIP + FER model · Train RAF-DB, mean over five test sets | mean accuracy: 55.73 -> 63.48 | Improved |
| add CAFE to different backbones (ablation) | MobileNet · Train RAF-DB, mean over five test sets | mean accuracy: 54.66 -> 56.30 | Improved |
| add CAFE to ResNet-50 (ablation) | ResNet-50 · Train RAF-DB, mean over five test sets | mean accuracy: 60.98 -> 64.07 | Improved |
| stronger source-domain FER methods (EAC, OFER) versus older SCN | FER baselines · Train RAF-DB, test FERPlus | accuracy: 58.37 (SCN) -> 54.38 (EAC), 53.90 (OFER) | Hurt |
Closest papers by meaning
- Open-Set Facial Expression Recognition (AAAI 2024 · similarity 0.90)
- Cross-Domain Facial Expression Recognition: A Unified Evaluation Benchmark and Adversarial Graph Learning (TPAMI 2021 · similarity 0.90)
- FERV39k: A Large-Scale Multi-Scene Dataset for Facial Expression Recognition in Videos (CVPR 2022 · similarity 0.88)
- Personalized Feature Translation for Expression Recognition: An Efficient Source-Free Domain Adaptation Method (ICLR 2026 · similarity 0.88)
- FAIR-FER: A Latent Alignment Approach for Mitigating Bias in Facial Expression Recognition (Student Abstract) (AAAI 2024 · similarity 0.86)
- PrefAce: Face-Centric Pretraining with Self-Structure Aware Distillation (AAAI 2024 · 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.
