🚀 News

• [2026/03/07] :fire: :fire: :fire: We launched the 3rd Micro-Action Analysis Grand Challenge (MAC 2026) associated with ACM Multimedia 2026.
• [2026/02/21] :fire: :fire: :fire: Our benchmark MA-Bench about fine-grained micro-action understanding via MLLMs is accepted by CVPR 2026.
• [2025/07/06] Our paper MMN about Skeleton-based Micro-Action Recognition is accepted by ACM MM 2025.
• [2025/06/26] Our dataset MMA-52 about Multi-label Micro-Action Detection is accepted by ICCV 2025.
• [2025/06/10] We launched the 2nd Micro-Action Analysis Grand Challenge (MAC 2025) associated with ACM Multimedia 2025.
• [2024/12/15] Our paper PCAN about Micro-Action Recognition is accepted by AAAI 2025.
• [2024/07/09] We released the MMA-52 dataset for the Multi-label Micro-Action Detection task. Report
• [2024/04/26] We launched the Micro-Action Analysis Grand Challenge (MAC 2024) associated with ACM Multimedia 2024.
• [2024/04/16] We released the source code of MANet.

---

📘 Introduction

Micro-action is an imperceptible non-verbal behaviour characterised by low-intensity movement. It offers insights into the feelings and intentions of individuals and is important for human-oriented applications such as emotion recognition and psychological assessment. However, the identification, differentiation, and understanding of micro-actions pose challenges due to the imperceptible and inaccessible nature of these subtle human behaviors in everyday life. In this study, we innovatively collect a new micro-action dataset designated as Micro-action-52 (MA-52), and propose a benchmark named micro-action network (MANet) for micro-action recognition (MAR) task. Uniquely, MA-52 provides the whole-body perspective including gestures, upper- and lower-limb movements, attempting to reveal comprehensive micro-action cues. In detail, MA-52 contains 52 micro-action categories along with seven body part labels, and encompasses a full array of realistic and natural micro-actions, accounting for 205 participants and 22,422 video instances collated from the psychological interviews. Based on the proposed dataset, we assess MANet and other nine prevalent action recognition methods. MANet incorporates squeeze-and-excitation (SE) and temporal shift module (TSM) into the ResNet architecture for modeling the spatiotemporal characteristics of micro-actions. Then, a joint-embedding loss is designed for semantic matching between video and action labels; the loss is used to better distinguish between visually similar yet distinct micro-action categories. The extended application in emotion recognition has demonstrated one of the important values of our proposed dataset and method. In the future, further exploration of human behaviour, emotion, and psychological assessment will be conducted in depth.

---

📂 Data

Download

The datasets are only to be used for non-commercial scientific purposes. You may request access to the dataset by completing the Google Form provided and corresponding LA files. We will respond promptly upon receipt of your application. If you have difficulty in filling out the form, we can also accept the application by [email].

Micro-Action-52 dataset (MA-52):

Multi-label Micro-Action-52 dataset (MMA-52):

MA-52 Statistics

Micro-Action examples

For more micro-action samples, please refer to MA-52 Dataset Samples.zip in huggingface.

A1: shaking body	A2: turning around	A3: sitting straightly
B1 nodding	B2 shaking head	B3 turning head

📄 Citation

Please consider citing the related paper in your publications if it helps your research.

@article{guo2024benchmarking,
  title={Benchmarking Micro-action Recognition: Dataset, Methods, and Applications},
  author={Guo, Dan and Li, Kun and Hu, Bin and Zhang, Yan and Wang, Meng},
  journal={IEEE Transactions on Circuits and Systems for Video Technology},
  year={2024},
  volume={34},
  number={7},
  pages={6238-6252}
}

@article{li2024mmad,
  title={Mmad: Multi-label micro-action detection in videos},
  author={Li, Kun and Liu, Pengyu and Guo, Dan and Wang, Fei and Wu, Zhiliang and Fan, Hehe and Wang, Meng},
  journal={arXiv preprint arXiv:2407.05311},
  year={2024}
}

@inproceedings{li2025prototypical,
  title={Prototypical calibrating ambiguous samples for micro-action recognition},
  author={Li, Kun and Guo, Dan and Chen, Guoliang and Fan, Chunxiao and Xu, Jingyuan and Wu, Zhiliang and Fan, Hehe and Wang, Meng},
  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
  volume={39},
  number={5},
  pages={4815--4823},
  year={2025}
}


@misc{MicroAction2024,
  author       = {Guo, Dan and Li, Kun and Hu, Bin and Zhang, Yan and Wang, Meng},
  title        = {Micro-Action Benchmark},
  year         = {2024},
  howpublished = {\url{https://github.com/VUT-HFUT/Micro-Action}},
  note         = {Accessed: 2024-08-21}
}