Ramprasaath R. Selvaraju, Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh, & Dhruv Batra (2017)

Proceedings of the IEEE International Conference on Computer Vision, 618-626.

DOI: https://doi.org/10.1109/ICCV.2017.74

Abstract. Selvaraju et al.'s Grad-CAM technique — gradient-based class- activation maps for CNN visual explanations. The single most- cited visual-XAI method, frequently the explanation type users see when image-classification models are deployed with explanation overlays.

Tags: ai-usability xai computer-vision

Textbook of Usability

Grad-CAM; visual explanations from deep networks via gradient-based localization