A New Multi-Domain Benchmark for Micro-Action Recognition and Detection

2026-06-12

AI Summary

Existing micro-action analysis benchmarks are limited in scale, scene diversity, and evaluation protocols, hindering robust recognition and detection in complex real-world scenarios. To address this, this work introduces MMA-82, a large-scale, multi-domain micro-action benchmark encompassing four domains, 82 fine-grained action categories, and 77,856 annotated instances. The benchmark establishes two core tasks—recognition and multi-label detection—and incorporates cross-domain, few-shot, and zero-shot evaluation protocols. MMA-82 represents the first effort to enable large-scale, fine-grained, and multi-domain modeling of micro-actions, revealing their strong association with emotional states. Experimental results demonstrate that current methods struggle under domain shift, long-tailed distributions, and temporal localization challenges, while integrating micro-action cues significantly enhances emotion recognition performance.

Abstract

Micro-actions are short-duration, low-amplitude subtle body movements at the whole-body level that can reveal latent intentions, involuntary reactions, and fine-grained affective changes. Our previous MA-52 benchmark has provided an important foundation for micro-action recognition, but it remains limited in scale, scene diversity, task coverage, and evaluation protocols. To advance micro-action analysis toward more realistic and comprehensive settings, we introduce MMA-82, a large-scale multi-domain extension of MA-52. MMA-82 expands the label space from 52 to 82 fine-grained micro-action categories and covers four distinct domains, including laboratory interviews, street interviews, psychiatric patient interviews, and emotion-rich television videos, resulting in 77,856 annotated instances from 454 subjects. Built upon MMA-82, we establish two core tasks: Micro-Action Recognition and Multi-label Micro-Action Detection. For recognition, we further define in-domain and cross-domain protocols, including few-shot and zero-shot settings, to evaluate model robustness, transferability, and generalization. Extensive experiments show that current methods still struggle with realistic micro-action understanding, especially under domain shift, long-tailed category distributions, and complex temporal localization. Beyond benchmarking, we investigate the relationship between micro-actions and emotion, showing that micro-actions are strongly associated with emotional states and provide complementary cues to facial micro-expressions for improved emotion recognition. These results demonstrate that MMA-82 serves as a comprehensive and challenging benchmark for realistic micro-action analysis and a valuable resource for human-centered AI. MMA-82 is available at https://github.com/LpyNow/MMA-82.

Problem

• micro-action recognition
• micro-action detection
• multi-domain benchmark
• domain shift
• fine-grained action analysis

Innovation

• micro-action recognition
• multi-domain benchmark
• multi-label detection
• cross-domain generalization
• emotion recognition