Generalizing GradCAM for Embedding Networks
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 15 of its 21 references that are in the map. First posted on arXiv 1 Feb 2024.
Builds on
15 of 21 references.
References · 15
- 2015FaceNet: A Unified Embedding for Face Recognition and Clusteringcited by 2,833
- 2015Learning Deep Features for Discriminative Localizationcited by 2,430
- 2013Visualizing and Understanding Convolutional Networkscited by 2,278
- 2019CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparisoncited by 1,410
- 2017Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networkscited by 936
- 2017Sampling Matters in Deep Embedding Learningcited by 285
- 2017No Fuss Distance Metric Learning using Proxiescited by 225
- 2020Proxy Anchor Loss for Deep Metric Learningcited by 195
- 2020ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component Analysiscited by 107
- 2018Re-Identification with Consistent Attentive Siamese Networkscited by 40
- 2019U-CAM: Visual Explanation using Uncertainty based Class Activation Mapscited by 36
- 2021F-CAM: Full Resolution Class Activation Maps via Guided Parametric Upscalingcited by 26
3 more
- 2019Visualizing Deep Similarity Networkscited by 24
- 2020Adapting Grad-CAM for Embedding Networkscited by 21
- 2015Pose Embeddings: A Deep Architecture for Learning to Match Human Posescited by 3
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.