Research map / Label-efficient and robust learning / Recognition and long-tail
When Facial Expression Recognition Meets Few-Shot Learning: A Joint and Alternate Learning Framework
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 emotion branch and joint multi-task learning to the similarity branch (ablation) | EGS-Net · CFEE_C, EmotioNet_C | 5-shot accuracy: 67.05 -> 68.38; 58.83 -> 60.52 | Helped |
| Add alternate learning between the emotion and similarity branches (ablation) | EGS-Net · CFEE_C, EmotioNet_C | 5-shot accuracy: 67.33 -> 68.38; 59.25 -> 60.52 | Helped |
| Add alternate learning to the emotion branch (ablation) | Emotion branch · CFEE_C, EmotioNet_C | 5-shot accuracy: 65.46 -> 66.77; 57.44 -> 59.25 | Helped |
| Add weight decay during alternate learning (ablation) | EGS-Net · CFEE, CFEE_C, EmotioNet, EmotioNet_C | 5-shot accuracy: 71.31 -> 72.17; 67.03 -> 68.38; 59.33 -> 59.77; 59.73 -> 60.52 | Helped |
| Train with multiple basic-expression source domains instead of one (ablation) | Emotion and similarity branches · CFEE and EmotioNet datasets and subsets | accuracy: S_b (single) -> S_b (multiple): CFEE 65.41 -> 69.69 (5-shot), EmotioNet 56.35 -> 57.49 (5-shot) | Helped |
| Use an episodically trained similarity branch instead of the emotion branch for unseen compound-expression classification (ablation) | Independent branches · CFEE_C and EmotioNet_C | accuracy: 62.69 vs 55.00 (CFEE_C 5-shot); 57.66 vs 54.76 (EmotioNet_C 5-shot) | Helped |
| Use the graph-convolution metric module instead of the ProtoNet metric module within EGS-Net (ablation) | EGS-Net · CFEE and EmotioNet | 5-shot accuracy: 72.17 -> 73.79; 59.77 -> 62.12 | Helped |
| Add the EGS-Net framework to the ProtoNet metric baseline | ProtoNet · CFEE and EmotioNet | 5-shot accuracy: 69.69 -> 72.17; 57.49 -> 59.77 | Helped |
Closest papers by meaning
- Facial-R1: Aligning Reasoning and Recognition for Facial Emotion Analysis (AAAI 2026 · similarity 0.87)
- Open-Set Facial Expression Recognition (AAAI 2024 · similarity 0.87)
- Navigating Label Ambiguity for Facial Expression Recognition in the Wild (AAAI 2025 · similarity 0.87)
- QCS:Feature Refining from Quadruplet Cross Similarity for Facial Expression Recognition (AAAI 2025 · similarity 0.86)
- A Unified and Interpretable Emotion Representation and Expression Generation (CVPR 2024 · similarity 0.86)
- Lifting Scheme-Based Implicit Disentanglement of Emotion-Related Facial Dynamics in the Wild (AAAI 2025 · 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.