arXiv:2606.14096v1 Announce Type: new 
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.

تم تقديم معيار جديد، MMA-82، لتعزيز التعرف على الميكرو-أفعال واكتشافها، مما يوسع الإطار السابق MA-52. يتضمن هذا المعيار الجديد 77,856 حالة موضوعة عبر 82 فئة دقيقة من الميكرو-أفعال وأربعة مجالات متميزة، بما في ذلك المقابلات ومقاطع الفيديو الغنية بالعواطف.

Se ha introducido un nuevo benchmark, MMA-82, para mejorar el reconocimiento y la detección de microacciones, ampliando el marco anterior MA-52. Este nuevo conjunto de datos incluye 77,856 instancias anotadas en 82 categorías de microacciones y cuatro dominios distintos, incluidas entrevistas y videos ricos en emociones.

Un nouveau benchmark, MMA-82, a été introduit pour améliorer la reconnaissance et la détection des micro-actions, élargissant le cadre précédent MA-52. Ce nouvel ensemble de données comprend 77 856 instances annotées réparties sur 82 catégories de micro-actions et quatre domaines distincts, y compris des interviews et des vidéos riches en émotions.

A new benchmark, MMA-82, has been introduced to enhance micro-action recognition and detection, expanding the previous MA-52 framework. This new dataset includes 77,856 annotated instances across 82 micro-action categories and four distinct domains, including interviews and emotion-rich videos.