Region-aware Image-based Human Action Retrieval with Transformers
This addresses the neglected problem of image-based action retrieval for computer vision applications, though it appears incremental as it builds on existing datasets and transformer architectures.
The paper tackles image-based human action retrieval by developing an end-to-end model that learns action representations from the anchored person, contextual regions, and global image, using a novel fusion transformer module to integrate these features. Experiments on Stanford-40 and PASCAL VOC 2012 Action datasets show it significantly outperforms previous approaches.
Human action understanding is a fundamental and challenging task in computer vision. Although there exists tremendous research on this area, most works focus on action recognition, while action retrieval has received less attention. In this paper, we focus on the neglected but important task of image-based action retrieval which aims to find images that depict the same action as a query image. We establish benchmarks for this task and set up important baseline methods for fair comparison. We present an end-to-end model that learns rich action representations from three aspects: the anchored person, contextual regions, and the global image. A novel fusion transformer module is designed to model the relationships among different features and effectively fuse them into an action representation. Experiments on the Stanford-40 and PASCAL VOC 2012 Action datasets show that the proposed method significantly outperforms previous approaches for image-based action retrieval.