Hierarchical Explanations for Video Action Recognition

Download PDF
Open Webpage

Sadaf Gulshad, Teng Long, Nanne van Noord

Published: 01 Jan 2023, Last Modified: 24 Sept 2025XAI4CV 2023EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: To interpret deep neural networks, one main approach is to dissect the visual input and find the prototypical parts responsible for the classification. However, existing methods often ignore the hierarchical relationship between these prototypes, and thus can not explain semantic concepts at both higher level (e.g., water sports) and lower level (e.g., swimming). In this paper inspired by human cognition system, we leverage hierarchal information to deal with uncertainty. To this end, we propose HIerarchical Prototypical Explainer (HIPE) to build hierarchical relations between prototypes and classes. The faithfulness of our method is verified by reducing accuracy-explainability trade-off on UCF-101 while providing multi-level explanations.