Object-ABN: Learning to Generate Sharp Attention Maps for Action Recognition

Abstract

In this paper we propose an extension of the Attention Branch Network (ABN) by using instance segmentation for generating sharper attention maps for action recognition. Methods for visual explanation such as Grad-CAM usually generate blurry maps which are not intuitive for humans to understand, particularly in recognizing actions of people in videos. Our proposed method, Object-ABN, tackles this issue by introducing a new mask loss that makes the generated attention maps close to the instance segmentation result. Further the PC loss and multiple attention maps are introduced to enhance the sharpness of the maps and improve the performance of classification. Experimental results with UCF101 and SSv2 shows that the generated maps by the proposed method are much clearer qualitatively and quantitatively than those of the original ABN.

9 pages

References in corpus (8)

• UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild
• The Kinetics Human Action Video Dataset
• Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Survey
• Attentional Pooling for Action Recognition
• A Comprehensive Study of Deep Video Action Recognition
• Video Action Understanding
• GTA: Global Temporal Attention for Video Action Understanding
• Spatio-Temporal Perturbations for Video Attribution