Grad-CAM++: Generalized Gradient-Based Visual Explanations for Deep Convolutional Networks

Chattopadhay, Aditya and Sarkar, Anirban and Howlader, Prantik and Balasubramanian, Vineeth N (2018) Grad-CAM++: Generalized Gradient-Based Visual Explanations for Deep Convolutional Networks. In: IEEE Winter Conference on Applications of Computer Vision (WACV), 12-15 March 2018, USA.

Full text not available from this repository. (Request a copy)

Abstract

Over the last decade, Convolutional Neural Network (CNN) models have been highly successful in solving complex vision based problems. However, deep models are perceived as "black box" methods considering the lack of understanding of their internal functioning. There has been a significant recent interest to develop explainable deep learning models, and this paper is an effort in this direction. Building on a recently proposed method called Grad-CAM, we propose Grad-CAM++ to provide better visual explanations of CNN model predictions (when compared to Grad-CAM), in terms of better localization of objects as well as explaining occurrences of multiple objects of a class in a single image. We provide a mathematical explanation for the proposed method, Grad-CAM++, which uses a weighted combination of the positive partial derivatives of the last convolutional layer feature maps with respect to a specific class score as weights to generate a visual explanation for the class label under consideration. Our extensive experiments and evaluations, both subjective and objective, on standard datasets showed that Grad-CAM++ indeed provides better visual explanations for a given CNN architecture when compared to Grad-CAM.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Balasubramanian, Vineeth NUNSPECIFIED
Item Type: Conference or Workshop Item (Paper)
Subjects: Computer science
Divisions: Department of Computer Science & Engineering
Depositing User: Team Library
Date Deposited: 24 Jan 2019 10:51
Last Modified: 24 Jan 2019 10:51
URI: http://raiithold.iith.ac.in/id/eprint/4757
Publisher URL: http://doi.org/10.1109/WACV.2018.00097
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 4757 Statistics for this ePrint Item