Keep CALM and Improve Visual Feature Attribution

TL;DR

This paper introduces CALM, an improved class activation mapping method that integrates attribution into the training process, leading to more accurate visual feature attribution and better weakly-supervised localization.

Contribution

CALM explicitly incorporates attribution into the training graph using latent variables and EM, enhancing interpretability and localization performance over CAM.

Findings

CALM outperforms CAM in identifying discriminative attributes.

CALM achieves better results on weakly-supervised object localization benchmarks.

The method improves attribution accuracy and interpretability.

Abstract

The class activation mapping, or CAM, has been the cornerstone of feature attribution methods for multiple vision tasks. Its simplicity and effectiveness have led to wide applications in the explanation of visual predictions and weakly-supervised localization tasks. However, CAM has its own shortcomings. The computation of attribution maps relies on ad-hoc calibration steps that are not part of the training computational graph, making it difficult for us to understand the real meaning of the attribution values. In this paper, we improve CAM by explicitly incorporating a latent variable encoding the location of the cue for recognition in the formulation, thereby subsuming the attribution map into the training computational graph. The resulting model, class activation latent mapping, or CALM, is trained with the expectation-maximization algorithm. Our experiments show that CALM identifies discriminative attributes for image classifiers more accurately than CAM and other visual attribution baselines. CALM also shows performance improvements over prior arts on the weakly-supervised object localization benchmarks. Our code is available at https://github.com/naver-ai/calm.