R
Ramprasaath R. Selvaraju
Researcher at Georgia Institute of Technology
Publications - 29
Citations - 11915
Ramprasaath R. Selvaraju is an academic researcher from Georgia Institute of Technology. The author has contributed to research in topics: Convolutional neural network & Context (language use). The author has an hindex of 15, co-authored 26 publications receiving 6867 citations. Previous affiliations of Ramprasaath R. Selvaraju include Salesforce.com & Virginia Tech.
Papers
More filters
Proceedings ArticleDOI
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Ramprasaath R. Selvaraju,Michael Cogswell,Abhishek Das,Ramakrishna Vedantam,Devi Parikh,Dhruv Batra +5 more
TL;DR: This work combines existing fine-grained visualizations to create a high-resolution class-discriminative visualization, Guided Grad-CAM, and applies it to image classification, image captioning, and visual question answering (VQA) models, including ResNet-based architectures.
Journal ArticleDOI
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Ramprasaath R. Selvaraju,Michael Cogswell,Abhishek Das,Ramakrishna Vedantam,Devi Parikh,Devi Parikh,Dhruv Batra,Dhruv Batra +7 more
TL;DR: Grad-CAM as mentioned in this paper uses the gradients of any target concept (e.g., a dog in a classification network or a sequence of words in captioning network) flowing into the final convolutional layer to produce a coarse localization map highlighting the important regions in the image for predicting the concept.
Grad-CAM: Why did you say that? Visual Explanations from Deep Networks via Gradient-based Localization
Ramprasaath R. Selvaraju,Abhishek Das,Ramakrishna Vedantam,Michael Cogswell,Devi Parikh,Dhruv Batra +5 more
TL;DR: It is shown that Guided Grad-CAM helps untrained users successfully discern a "stronger" deep network from a "weaker" one even when both networks make identical predictions, and also exposes the somewhat surprising insight that common CNN + LSTM models can be good at localizing discriminative input image regions despite not being trained on grounded image-text pairs.
Posted Content
Grad-CAM: Why did you say that? Visual Explanations from Deep Networks via Gradient-based Localization
Ramprasaath R. Selvaraju,Abhishek Das,Ramakrishna Vedantam,Michael Cogswell,Devi Parikh,Dhruv Batra +5 more
TL;DR: Grad-CAM as discussed by the authors uses the class-specific gradient information flowing into the final convolutional layer of a CNN to produce a coarse localization map of the important regions in the image.
Posted Content
Grad-CAM: Why did you say that?
Ramprasaath R. Selvaraju,Abhishek Das,Ramakrishna Vedantam,Michael Cogswell,Devi Parikh,Dhruv Batra +5 more
TL;DR: This work proposes a technique for making Convolutional Neural Network (CNN)-based models more transparent by visualizing input regions that are 'important' for predictions -- or visual explanations, called Gradient-weighted Class Activation Mapping (Grad-CAM), which uses class-specific gradient information to localize important regions.