Describing Common Human Visual Actions in Images
This work addresses the need for a data-driven, exhaustive action dataset for computer vision researchers, though it is incremental as it builds on existing datasets like MS COCO.
The authors tackled the problem of identifying common human actions in still images by analyzing the MS COCO dataset, resulting in a new dataset called COCO-a with 140 annotated visual actions and localized subjects and objects. They provided statistical analysis of annotation accuracy and action combinations.
Which common human actions and interactions are recognizable in monocular still images? Which involve objects and/or other people? How many is a person performing at a time? We address these questions by exploring the actions and interactions that are detectable in the images of the MS COCO dataset. We make two main contributions. First, a list of 140 common `visual actions', obtained by analyzing the largest on-line verb lexicon currently available for English (VerbNet) and human sentences used to describe images in MS COCO. Second, a complete set of annotations for those `visual actions', composed of subject-object and associated verb, which we call COCO-a (a for `actions'). COCO-a is larger than existing action datasets in terms of number of actions and instances of these actions, and is unique because it is data-driven, rather than experimenter-biased. Other unique features are that it is exhaustive, and that all subjects and objects are localized. A statistical analysis of the accuracy of our annotations and of each action, interaction and subject-object combination is provided.