mlsanityData to 7 Oct 2026

B-MoE: A Body-Part-Aware Mixture-of-Experts "All Parts Matter" Approach to Micro-Action Recognition

arXiv 2603.24245 ↗First posted 25 Mar 2026cs.CVcited by 0

On the map: Mixture-of-experts → MoE in reinforcement learning and robotics

Abstract (arXiv)

Micro-actions, fleeting and low-amplitude motions, such as glances, nods, or minor posture shifts, carry rich social meaning but remain difficult for current action recognition models to recognize due to their subtlety, short duration, and high inter-class ambiguity. In this paper, we introduce B-MoE, a Body-part-aware Mixture-of-Experts framework designed to explicitly model the structured nature of human motion. In B-MoE, each expert specializes in a distinct body region (head, body, upper limbs, lower limbs), and is based on the lightweight Macro-Micro Motion Encoder (M3E) that captures long-range contextual structure and fine-grained local motion. A cross-attention routing mechanism learns inter-region relationships and dynamically selects the most informative regions for each micro-action. B-MoE uses a dual-stream encoder that fuses these region-specific semantic cues with global motion features to jointly capture spatially localized cues and temporally subtle variations that characterize micro-actions. Experiments on three challenging benchmarks (MA-52, SocialGesture, and MPII-GroupInteraction) show consistent state-of-theart gains, with improvements in ambiguous, underrepresented, and low amplitude classes.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in MoE in reinforcement learning and robotics, one of 12 directions in Mixture-of-experts. Builds on 28 of its 36 references that are in the map. First posted on arXiv 25 Mar 2026.

Builds on

28 of 36 references.

References · 28

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,259
  3. 2021Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsitycited by 3,014
  4. 2018SlowFast Networks for Video Recognitioncited by 1,862
  5. 2021Is Space-Time Attention All You Need for Video Understanding?cited by 1,609
  6. 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,199
  7. 2021Video Swin Transformercited by 1,090
  8. 2016Temporal Segment Networks: Towards Good Practices for Deep Action Recognitioncited by 899
  9. 2018TSM: Temporal Shift Module for Efficient Video Understandingcited by 852
  10. 2023VideoMAE V2: Scaling Video Masked Autoencoders with Dual Maskingcited by 516
  11. 2021Revisiting Skeleton-based Action Recognitioncited by 256
  12. 2023From Sparse to Soft Mixtures of Expertscited by 245
16 more
  1. 2024Sapiens: Foundation for Human Vision Modelscited by 229
  2. 2022UniFormer: Unifying Convolution and Self-attention for Visual Recognitioncited by 206
  3. 2023TriDet: Temporal Action Detection with Relative Boundary Modelingcited by 141
  4. 2025A Comprehensive Survey of Mixture-of-Experts: Algorithms, Theory, and Applicationscited by 91
  5. 2021iMiGUE: An Identity-free Video Dataset for Micro-Gesture Understanding and Emotion Analysiscited by 63
  6. 2024Benchmarking Micro-action Recognition: Dataset, Methods, and Applicationscited by 48
  7. 2020Temporal Interlacing Networkcited by 45
  8. 2024Mixture of Nested Experts: Adaptive Processing of Visual Tokenscited by 25
  9. 2018Detecting Low Rapport During Natural Interactions in Small Groups from Non-Verbal Behaviourcited by 24
  10. 2024Prototypical Calibrating Ambiguous Samples for Micro-Action Recognitioncited by 22
  11. 2024MMAD: Multi-label Micro-Action Detection in Videoscited by 20
  12. 2022Bodily Behaviors in Social Interaction: Novel Annotations and State-of-the-Art Evaluationcited by 19
  13. 2022Sparse Mixers: Combining MoE and Mixing to build a more efficient BERTcited by 12
  14. 2025SocialGesture: Delving into Multi-person Gesture Understandingcited by 11
  15. 2025Mixture of Experts Guided by Gaussian Splatters Matters: A new Approach to Weakly-Supervised Video Anomaly Detectioncited by 6
  16. 2025MixANT: Observation-dependent Memory Propagation for Stochastic Dense Action Anticipationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Spatiotemporal-Untrammelled Mixture of Experts for Multi-Person Motion Prediction
  2. 2026Fine-Grained Action Recognition with Cross-Attentive Latent Sparse Experts
  3. 2026SemanticFace: Semantic Facial Action Estimation via Semantic Distillation in Interpretable Space
  4. 2025PCBEAR: Pose Concept Bottleneck for Explainable Action Recognition
  5. 2026MAC 2026: Advancing Micro-Action Analysis Towards Fine-Grained Understanding
  6. 2026MA-Bench: Towards Fine-grained Micro-Action Understanding
  7. 2026Recognition-Conditioned Reasoning: A Training-Free Multimodal-LLM Pipeline for Fine-Grained Micro-Action Understanding
  8. 2025DEFT-LLM: Disentangled Expert Feature Tuning for Micro-Expression Recognition