iMiGUE-3K is the largest in-the-wild micro-gesture video dataset with 3.4K clips and 37M frames from real interviews, supporting self-supervised foundation models and benchmarks that show micro-gestures improve emotion understanding.
Mm-gesture: towards precise micro-gesture recognition through multimodal fusion
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
fields
cs.CV 7years
2026 7verdicts
UNVERDICTED 7representative citing papers
A decoupled adapter with independent spatial-temporal branches via depthwise convolutions and a dynamic augmentation strategy for long-tail data achieves first place with F1 0.43808 in a micro-gesture recognition challenge.
Micro-DualNet employs dual ST and TS pathways with entity-level adaptive routing and Mutual Action Consistency loss to achieve competitive results on MA-52 and state-of-the-art on iMiGUE for micro-action recognition.
A multi-modal system combining skeleton/heatmap/RGB features with cross-modal pseudo-labeling and semantic losses achieves 68.13% F1-score and 4th place on the MiGA-IJCAI micro-gesture challenge.
DyFADet+ extends a prior detector with gated RGB-skeleton fusion and reports 40.88 F1 on the SMG dataset for micro-gesture online recognition.
Ensemble of self-supervised RGB model and supervised models achieves new SOTA of 74.419% on iMiGUE micro-gesture dataset.
A competition-winning multi-modal model for hidden emotion recognition integrates static and dynamic pose features via cross-attention and MIL pooling while noting representation collapse in vision foundation models on micro-dynamic tasks.
citing papers explorer
-
iMiGUE-3K: A Large-Scale Benchmark for Micro-Gesture Analysis with Self-Supervised Learning
iMiGUE-3K is the largest in-the-wild micro-gesture video dataset with 3.4K clips and 37M frames from real interviews, supporting self-supervised foundation models and benchmarks that show micro-gestures improve emotion understanding.
-
Spatial-Temporal Decoupled Adapter for Micro-gesture Online Recognition
A decoupled adapter with independent spatial-temporal branches via depthwise convolutions and a dynamic augmentation strategy for long-tail data achieves first place with F1 0.43808 in a micro-gesture recognition challenge.
-
Micro-DualNet: Dual-Path Spatio-Temporal Network for Micro-Action Recognition
Micro-DualNet employs dual ST and TS pathways with entity-level adaptive routing and Mutual Action Consistency loss to achieve competitive results on MA-52 and state-of-the-art on iMiGUE for micro-action recognition.
-
A Multi-Modal Framework with Cross-Subject Pseudo-Labeling and Semantic Alignment for Micro-Gesture Recognition
A multi-modal system combining skeleton/heatmap/RGB features with cross-modal pseudo-labeling and semantic losses achieves 68.13% F1-score and 4th place on the MiGA-IJCAI micro-gesture challenge.
-
Motion Reinforces Appearance: RGB-Skeleton Gated Residual Fusion for Micro-Gesture Online Recognition
DyFADet+ extends a prior detector with gated RGB-skeleton fusion and reports 40.88 F1 on the SMG dataset for micro-gesture online recognition.
-
Self-supervised Learning Matters: A Simple Ensemble Solution for Micro-Gesture Recognition
Ensemble of self-supervised RGB model and supervised models achieves new SOTA of 74.419% on iMiGUE micro-gesture dataset.
-
Rethinking the Role of Feature Engineering and Learning Strategies in Few-Shot Hidden Emotion Recognition
A competition-winning multi-modal model for hidden emotion recognition integrates static and dynamic pose features via cross-attention and MIL pooling while noting representation collapse in vision foundation models on micro-dynamic tasks.