English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Survey: Class Activation Mapping in Explainable Computer Vision — A Method-Centered Review (arXiv 2508.03414)

Forum topic · 小凯 · 2026-08-14

Summary

A new survey by AmirHossein Eshghi, Hamid Saadatfar, and Seyyed Ali Hoseini reviews the evolution of Class Activation Mapping (CAM), one of the most widely used visual explanation families in explainable AI. Covering 57 method-focused papers published since 2016, the authors build a taxonomy that organizes CAM approaches by attribution mechanism, architectural dependency, and evaluation target. The survey traces the field's expansion from the original global-average-pooling CNN classifiers to gradient-based and gradient-free explanations, ablation methods, high-resolution upsampling, weakly supervised localization and segmentation, transformer token attribution, causal and debiasing techniques, and foundation-model-era approaches using CLIP, DINO, and SAM. The authors identify a clear trend: the field is moving from explaining a single class score in a single-resolution CNN layer toward comparative, multi-layer, probabilistic, token-aware, and foundation-model-aware explanations. However, evaluation remains fragmented, with faithfulness, localization, robustness, computational cost, and human trust measured under inconsistent protocols. The paper, posted on arXiv as 2508.03414, highlights both each method's contributions and the gaps it leaves for successors.

Paper Overview

  • Field: Computer Vision
  • Authors: AmirHossein Eshghi, Hamid Saadatfar, Seyyed Ali Hoseini
  • Published: 2026-08-13
  • arXiv: 2508.03414
  • Key Points

    Class activation mapping (CAM) is one of the most widely used visual explanation families in explainable AI. Its goal is intuitive: convert internal model evidence into heatmaps that highlight image regions, convolution channels, tokens, or patches supporting a target class or concept.

    Since the first CAM formulation in 2016, the field has expanded far beyond global-average-pooling CNN classifiers. CAM-style methods now include:

  • Gradient-based post-hoc explanations
  • Gradient-free scores and ablation methods
  • High-resolution upsampling techniques
  • Weakly supervised localization and segmentation
  • Transformer token attribution
  • Causal and debiasing approaches
  • Foundation-model-era methods using CLIP, DINO, SAM, or feature-distribution comparison

Survey Structure

The survey synthesizes 57 method-focused studies published since 2016. The authors:

1. Develop a taxonomy that separates methods by attribution mechanism, architectural dependency, and evaluation target 2. Review gradient-based CAM, recent and hybrid CAM-style methods, and model- or architecture-aware approaches

Trends and Open Problems

A dominant trend is clear: the field is shifting from explaining a single class score in a single-resolution CNN layer toward comparative, multi-layer, probabilistic, token-aware, and foundation-model-aware explanations.

At the same time, evaluation remains fragmented — faithfulness, localization, robustness, computational cost, and human trust are typically measured under different protocols. The survey therefore emphasizes not only each method's contributions, but also the gaps it leaves and how subsequent methods attempt to fill them.

--- *Auto-collected on 2026-08-14*

Tags

#class-activation-mapping#explainable-ai#computer-vision#survey#arxiv#transformer-attribution#foundation-models#interpretability

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178633457