Hierarchical Explanations for Video Action Recognition

Sadaf Gulshad, Teng Long, Nanne van Noord; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2023, pp. 3702-3707

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.

Related Material


[pdf] [arXiv]
[bibtex]
@InProceedings{Gulshad_2023_CVPR, author = {Gulshad, Sadaf and Long, Teng and van Noord, Nanne}, title = {Hierarchical Explanations for Video Action Recognition}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {3702-3707} }