Generalizing GradCAM for Embedding Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Bachhawat, Mudit
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929234968051712
author Bachhawat, Mudit
author_facet Bachhawat, Mudit
contents Visualizing CNN is an important part in building trust and explaining model's prediction. Methods like CAM and GradCAM have been really successful in localizing area of the image responsible for the output but are only limited to classification models. In this paper, we present a new method EmbeddingCAM, which generalizes the Grad-CAM for embedding networks. We show that for classification networks, EmbeddingCAM reduces to GradCAM. We show the effectiveness of our method on CUB-200-2011 dataset and also present quantitative and qualitative analysis on the dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2402_00909
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Generalizing GradCAM for Embedding Networks
Bachhawat, Mudit
Computer Vision and Pattern Recognition
Visualizing CNN is an important part in building trust and explaining model's prediction. Methods like CAM and GradCAM have been really successful in localizing area of the image responsible for the output but are only limited to classification models. In this paper, we present a new method EmbeddingCAM, which generalizes the Grad-CAM for embedding networks. We show that for classification networks, EmbeddingCAM reduces to GradCAM. We show the effectiveness of our method on CUB-200-2011 dataset and also present quantitative and qualitative analysis on the dataset.
title Generalizing GradCAM for Embedding Networks
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.00909