AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions
This provides a new benchmark dataset for spatio-temporal action recognition in videos, addressing the problem of limited annotations in existing datasets for researchers in computer vision.
The paper introduces the AVA dataset, which densely annotates 80 atomic visual actions in 430 video clips with 1.58 million labels, and presents a novel action localization approach that achieves state-of-the-art results on existing datasets but only 15.6% mAP on AVA, highlighting the challenge of video understanding.
This paper introduces a video dataset of spatio-temporally localized Atomic Visual Actions (AVA). The AVA dataset densely annotates 80 atomic visual actions in 430 15-minute video clips, where actions are localized in space and time, resulting in 1.58M action labels with multiple labels per person occurring frequently. The key characteristics of our dataset are: (1) the definition of atomic visual actions, rather than composite actions; (2) precise spatio-temporal annotations with possibly multiple annotations for each person; (3) exhaustive annotation of these atomic actions over 15-minute video clips; (4) people temporally linked across consecutive segments; and (5) using movies to gather a varied set of action representations. This departs from existing datasets for spatio-temporal action recognition, which typically provide sparse annotations for composite actions in short video clips. We will release the dataset publicly. AVA, with its realistic scene and action complexity, exposes the intrinsic difficulty of action recognition. To benchmark this, we present a novel approach for action localization that builds upon the current state-of-the-art methods, and demonstrates better performance on JHMDB and UCF101-24 categories. While setting a new state of the art on existing datasets, the overall results on AVA are low at 15.6% mAP, underscoring the need for developing new approaches for video understanding.