Research map / Label-efficient and robust learning / Domain adaptation and generalization
Learn-to-Decompose: Cascaded Decomposition Network for Cross-Domain Few-Shot Facial Expression Recognition
Xinyi Zou, Yan Yan, Jing-Hao Xue, Si Chen, Hanzi Wang
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 |
|---|---|---|---|
| use partial regularization instead of full regularization (ablation) | CDNet · CFEE_C 1-shot (pre-trained) | accuracy: 55.91 -> 56.99 | Improved |
| use partial regularization instead of fixing the pre-trained decomposition block (ablation) | CDNet · RAF_C 5-shot (pre-trained) | accuracy: 62.12 -> 63.03 | Improved |
| add partial regularization from the pre-trained decomposition block (ablation) | CDNet · RAF_C 5-shot (pre-trained) | accuracy: 61.00 -> 63.03 | Improved |
| use sequential decomposition with shared LD modules instead of parallel modules (ablation) | CDNet · EmotioNet_C 1-shot (pre-trained) | accuracy: 54.27 -> 55.16 | Improved |
| use two-block LD module instead of single transformation module (ablation) | CDNet · RAF_C 1-shot (pre-trained) | accuracy: 43.22 -> 46.07 | Improved |
| add the cascaded decomposition design (no regularization) to ProtoNet baseline (ablation) | ProtoNet · CFEE_C 1-shot (pre-trained) | accuracy: 53.43 -> 55.17 | Improved |
| initialize fine-tuning from batch-trained pre-trained model (ablation) | CDNet · RAF_C 5-shot | accuracy: 61.75 -> 63.03 | Improved |
| initialize fine-tuning from batch-trained pre-trained model (ablation) | ProtoNet baseline · CFEE_C 1-shot | accuracy: 53.29 -> 53.43 | Improved |
| increase the number of cascaded LD modules beyond 3 (ablation) | CDNet · CFEE_C, EmotioNet_C, RAF_C | accuracy | Hurt |
| cascade 2 or 3 shared LD modules instead of 1 (ablation) | CDNet · CFEE_C, EmotioNet_C, RAF_C | accuracy | Improved |
| add a single LD module to the baseline (1 cascaded module vs 0) (ablation) | CDNet · CFEE_C, EmotioNet_C, RAF_C | accuracy | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Meta-Baseline — Meta-Baseline: Exploring Simple Meta-Learning for Few-Shot Learning (ICCV 2021) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- When Facial Expression Recognition Meets Few-Shot Learning: A Joint and Alternate Learning Framework (AAAI 2022 · similarity 0.91)
- Feature Decomposition and Reconstruction Learning for Effective Facial Expression Recognition (CVPR 2021 · similarity 0.89)
- Multi-Label Compound Expression Recognition: C-EXPR Database & Network (CVPR 2023 · similarity 0.89)
- Cross-Domain Facial Expression Recognition: A Unified Evaluation Benchmark and Adversarial Graph Learning (TPAMI 2021 · similarity 0.89)
- QCS:Feature Refining from Quadruplet Cross Similarity for Facial Expression Recognition (AAAI 2025 · similarity 0.86)
- Navigating Label Ambiguity for Facial Expression Recognition in the Wild (AAAI 2025 · similarity 0.84)
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.
