Research map / Label-efficient and robust learning / Recognition and long-tail
Context-Aware Emotion Recognition Networks
Jiyoung Lee, Seungryong Kim, Sunok Kim, Jungin Park, Kwanghoon Sohn
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 context stream (with context attention) to the face stream (ablation) | CAER-Net · CAER | accuracy: 74.13 -> 74.94 | Improved |
| add context stream without attention to the face stream (ablation) | CAER-Net · CAER | accuracy: 74.13 -> 74.36 | Improved |
| add context attention module (ablation) | CAER-Net · CAER | accuracy: 74.36 -> 74.94 | Improved |
| add adaptive fusion attention module (ablation) | CAER-Net · CAER | accuracy: 74.36 -> 75.57 | Improved |
| add fusion attention on top of context attention (ablation) | CAER-Net · CAER | accuracy: 74.94 -> 77.04 | Improved |
| add context attention on top of fusion attention (ablation) | CAER-Net · CAER | accuracy: 75.57 -> 77.04 | Improved |
| add face stream to context-only network (ablation) | CAER-Net · CAER | accuracy: 71.94 -> 74.94 | Improved |
| add context stream and both attention modules to face-only network (ablation) | CAER-Net-S · CAER-S | accuracy: 70.09 -> 73.51 | Improved |
| add face stream to context-only network (ablation) | CAER-Net-S · CAER-S | accuracy: 65.65 -> 73.51 | Improved |
| use temporal video input (dynamic 3D model) instead of static images | CAER-Net · CAER vs CAER-S | accuracy: 3.53 % improvement | Improved |
| replace fine-tuned ResNet with CAER-Net-S | CAER-Net-S · CAER-S | accuracy: 68.46 -> 73.51 | Improved |
| replace fine-tuned C3D with CAER-Net | CAER-Net · CAER | accuracy: 71.02 -> 77.04 | Improved |
Closest papers by meaning
- Context De-Confounded Emotion Recognition (CVPR 2023 · similarity 0.91)
- Towards Context-Aware Emotion Recognition Debiasing From a Causal Demystification Perspective via De-Confounded Training (TPAMI 2024 · similarity 0.89)
- Robust Emotion Recognition in Context Debiasing (CVPR 2024 · similarity 0.87)
- Emotion Recognition for Multiple Context Awareness (ECCV 2022 · similarity 0.87)
- CocoER: Aligning Multi-Level Feature by Competition and Coordination for Emotion Recognition (CVPR 2025 · similarity 0.86)
- Facial-R1: Aligning Reasoning and Recognition for Facial Emotion Analysis (AAAI 2026 · 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.
