Research map / Video / Action recognition
Rethinking Key-frame-based Micro-expression Recognition: A Robust and Accurate Framework Against Key-frame Errors
Zheyuan Zhang, Weihao Tang, Hong Chen
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 causal motion position learning module (CMPLM) (ablation) | CausalNet backbone · Composite ablation | UF1: 86.40 -> 87.01 | Improved |
| add causal motion position learning module (CMPLM) (ablation) | CausalNet backbone · Composite ablation | UAR: 86.55 -> 87.25 | Improved |
| add spatial-temporal causal attention (STCA) (ablation) | CausalNet w/ CMPLM · Composite ablation | UF1: 87.01 -> 88.42 | Improved |
| add spatial-temporal causal attention (STCA) (ablation) | CausalNet w/ CMPLM · Composite ablation | UAR: 87.25 -> 88.50 | Improved |
| add causal relation mining (CRM) (ablation) | CausalNet w/ CMPLM+STCA · Composite ablation | UF1: 88.42 -> 88.81 | Improved |
| add causal relation mining (CRM) (ablation) | CausalNet w/ CMPLM+STCA · Composite ablation | UAR: 88.50 -> 88.94 | Improved |
| add pseudo-attention (PA) (ablation) | CausalNet w/ CMPLM+STCA+CRM · Composite ablation | UF1: 88.81 -> 89.30 | Improved |
| add pseudo-attention (PA) (ablation) | CausalNet w/ CMPLM+STCA+CRM · Composite ablation | UAR: 88.94 -> 89.41 | Improved |
| use CausalNet instead of second-best method under key-frame error STD 10 | key-frame MER · Composite, key-frame error STD 10 frames | UF1: +2.70% | Improved |
| use CausalNet instead of second-best method under key-frame error STD 30 | key-frame MER · Composite, key-frame error STD 30 frames | UF1: +1.42% | Improved |
| use CausalNet instead of second-best method with automatic apex spotting | key-frame MER · CASME II automatic apex spotting | UAR: +2.62% UAR, +1.06% UF1 | Improved |
Closest papers by meaning
- Region-Aware Instance Consistency Learning for Micro-Expression Recognition (CVPR 2026 · similarity 0.87)
- From Pixels to Semantics: Unified Facial Action Representation Learning for Micro-Expression Analysis (ICLR 2026 · similarity 0.87)
- MOL: Joint Estimation of Micro-Expression, Optical Flow, and Landmark via Transformer-Graph-Style Convolution (TPAMI 2025 · similarity 0.86)
- FED-PsyAU: Privacy-Preserving Micro-Expression Recognition via Psychological AU Coordination and Dynamic Facial Motion Modeling (ICCV 2025 · similarity 0.86)
- Prototypical Calibrating Ambiguous Samples for Micro-Action Recognition (AAAI 2025 · similarity 0.85)
- MA-Bench: Towards Fine-grained Micro-Action Understanding (CVPR 2026 · similarity 0.83)
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.
