DecomCAM: Advancing Beyond Saliency Maps through Decomposition and Integration

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Yang, Yuguang, Guo, Runtang, Wu, Sheng, Wang, Yimi, Yang, Linlin, Fan, Bo, Zhong, Jilong, Zhang, Juan, Zhang, Baochang
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914815609405440
author Yang, Yuguang
Guo, Runtang
Wu, Sheng
Wang, Yimi
Yang, Linlin
Fan, Bo
Zhong, Jilong
Zhang, Juan
Zhang, Baochang
author_facet Yang, Yuguang
Guo, Runtang
Wu, Sheng
Wang, Yimi
Yang, Linlin
Fan, Bo
Zhong, Jilong
Zhang, Juan
Zhang, Baochang
contents Interpreting complex deep networks, notably pre-trained vision-language models (VLMs), is a formidable challenge. Current Class Activation Map (CAM) methods highlight regions revealing the model's decision-making basis but lack clear saliency maps and detailed interpretability. To bridge this gap, we propose DecomCAM, a novel decomposition-and-integration method that distills shared patterns from channel activation maps. Utilizing singular value decomposition, DecomCAM decomposes class-discriminative activation maps into orthogonal sub-saliency maps (OSSMs), which are then integrated together based on their contribution to the target concept. Extensive experiments on six benchmarks reveal that DecomCAM not only excels in locating accuracy but also achieves an optimizing balance between interpretability and computational efficiency. Further analysis unveils that OSSMs correlate with discernible object components, facilitating a granular understanding of the model's reasoning. This positions DecomCAM as a potential tool for fine-grained interpretation of advanced deep learning models. The code is avaible at https://github.com/CapricornGuang/DecomCAM.
format Preprint
id arxiv_https___arxiv_org_abs_2405_18882
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DecomCAM: Advancing Beyond Saliency Maps through Decomposition and Integration
Yang, Yuguang
Guo, Runtang
Wu, Sheng
Wang, Yimi
Yang, Linlin
Fan, Bo
Zhong, Jilong
Zhang, Juan
Zhang, Baochang
Computer Vision and Pattern Recognition
Interpreting complex deep networks, notably pre-trained vision-language models (VLMs), is a formidable challenge. Current Class Activation Map (CAM) methods highlight regions revealing the model's decision-making basis but lack clear saliency maps and detailed interpretability. To bridge this gap, we propose DecomCAM, a novel decomposition-and-integration method that distills shared patterns from channel activation maps. Utilizing singular value decomposition, DecomCAM decomposes class-discriminative activation maps into orthogonal sub-saliency maps (OSSMs), which are then integrated together based on their contribution to the target concept. Extensive experiments on six benchmarks reveal that DecomCAM not only excels in locating accuracy but also achieves an optimizing balance between interpretability and computational efficiency. Further analysis unveils that OSSMs correlate with discernible object components, facilitating a granular understanding of the model's reasoning. This positions DecomCAM as a potential tool for fine-grained interpretation of advanced deep learning models. The code is avaible at https://github.com/CapricornGuang/DecomCAM.
title DecomCAM: Advancing Beyond Saliency Maps through Decomposition and Integration
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.18882