Research map / Video / Action recognition
ARBEE: Towards Automated Recognition of Bodily Expression of Emotion in the Wild
Yu Luo, Jianbo Ye, Reginald B. Adams, Jia Li, Michelle G. Newman, James Z. 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 |
|---|---|---|---|
| initialize with Kinetics-pretrained model instead of ImageNet (ablation) | TSN · BoLD test set | ERS: 0.247 -> 0.245 | No clear effect |
| add faceless body (remove face) input (ablation) | TSN · BoLD test set | ERS: 0.247 -> 0.241 | Mixed |
| restrict input to face only (ablation) | TSN · BoLD test set | ERS: 0.247 -> 0.242 | Hurt |
| add whole body context to face-only input (ablation) | TSN · BoLD test set | mean average precision: 16.21 -> 17.02 | Improved |
| ensemble TSN-body with LMA features (ablation) | TSN-body · BoLD test set | ERS: 0.247 -> 0.249 | Improved |
| ensemble TSN-body with TSN-face (ablation) | TSN-body · BoLD test set | ERS: 0.247 -> 0.252 | Improved |
| ensemble TSN-body, TSN-face and LMA (ablation) | TSN-body · BoLD test set | ERS: 0.247 -> 0.253 | Improved |
| add LMA to TSN-body + TSN-face (ablation) | TSN-body + TSN-face · BoLD test set | mean average precision: 17.31 -> 17.14 | Hurt |
| use handcrafted LMA features instead of end-to-end ST-GCN | skeleton pipeline · BoLD test set | ERS: 0.194 -> 0.216 | Improved |
| use two-stream network instead of trajectory features | pixel pipeline · BoLD test set | ERS: 0.149 -> 0.247 | Improved |
Closest papers by meaning
- Understanding Emotional Body Expressions via Large Language Models (AAAI 2025 · similarity 0.88)
- FEEL: Quantifying Heterogeneity in Physiological Signals for Generalizable Emotion Recognition (NeurIPS 2025 · similarity 0.88)
- $E^3$: Exploring Embodied Emotion Through A Large-Scale Egocentric Video Dataset (NeurIPS 2024 · similarity 0.88)
- EmoNet-Face: An Expert-Annotated Benchmark for Synthetic Emotion Recognition (NeurIPS 2025 · similarity 0.86)
- Facial-R1: Aligning Reasoning and Recognition for Facial Emotion Analysis (AAAI 2026 · similarity 0.85)
- AffectGPT: A New Dataset, Model, and Benchmark for Emotion Understanding with Multimodal Large Language Models (ICML 2025 · 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.
