Research map / Label-efficient and robust learning / Recognition and long-tail
Label Distribution Learning on Auxiliary Label Space Graphs for Facial Expression Recognition
Check 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 |
|---|---|---|---|
| Use auxiliary-task label-space neighbors to guide label-distribution learning | LDL-ALSG with ResNet-50 · Overall average | accuracy (%): 69.33 -> 72.13 | Helped |
| Use auxiliary-task label-space neighbors to guide label-distribution learning | LDL-ALSG with ResNet-50 · Overall average, 15% label noise | accuracy (%): 65.12 -> 69.24 | Helped |
| Add facial landmark neighbors alongside action-unit neighbors (ablation) | LDL-ALSG · Overall average | accuracy (%): 72.01 -> 72.13 | Helped |
| Add action-unit neighbors alongside facial landmark neighbors (ablation) | LDL-ALSG · Overall average | accuracy (%): 71.87 -> 72.13 | Helped |
| Use facial landmark neighbors instead of action-unit neighbors (ablation) | LDL-ALSG · SFEW | accuracy (%): 56.50 -> 54.91 | Hurt |
| Increase AU neighbors from 2 to 4 (ablation) | LDL-ALSG · Overall average | accuracy (%): 71.98 -> 72.01 | No clear effect |
| Increase AU neighbors from 4 to 8 (ablation) | LDL-ALSG · Overall average | accuracy (%): 72.01 -> 71.69 | Hurt |
| Set λ to 0.0005 for guidance losses from both auxiliary tasks (ablation) | LDL-ALSG · Overall average | accuracy (%): 71.62 -> 72.13 | Helped |
| Set λ to 0.001 for guidance losses from both auxiliary tasks (ablation) | LDL-ALSG · Overall average | accuracy (%): 72.13 -> 71.62 | Hurt |
| Train with LDL-ALSG rather than the baseline | ResNet-18 FER model · Overall average | accuracy (%): 68.07 -> 70.08 | Helped |
Closest papers by meaning
- Facial Emotion Distribution Learning by Exploiting Low-Rank Label Correlations Locally (CVPR 2019 · similarity 0.89)
- Navigating Label Ambiguity for Facial Expression Recognition in the Wild (AAAI 2025 · similarity 0.87)
- Distribution Matching for Multi-Task Learning of Classification Tasks: A Large-Scale Study on Faces & Beyond (AAAI 2024 · similarity 0.86)
- GLDL: Graph Label Distribution Learning (AAAI 2024 · similarity 0.86)
- HGDL: Heterogeneous Graph Label Distribution Learning (NeurIPS 2024 · similarity 0.85)
- Open-Set Facial Expression Recognition (AAAI 2024 · similarity 0.85)
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.