AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

Grad-CAM: Visual Explanations From Deep Networks via Gradient-Based Localization

Ramprasaath R. Selvaraju, Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh, Dhruv Batra

ICCV 2017 · Vision-language and multimodal › Visual question answering · approach: Vision-language model VQA evaluation · 25,969 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Grad-CAM instead of Guided Backpropagation gradients for localizationVGG-16 · ILSVRC-15 val top-1 localization errortop-1 localization error: 61.12 -> 56.51Improved
use Grad-CAM on the unmodified VGG-16 instead of CAM with a GAP architectureVGG-16 · ILSVRC-15 valtop-1 classification error: 33.40 -> 30.38Improved
use Grad-CAM instead of CAM for top-5 localizationVGG-16 · ILSVRC-15 valtop-5 localization error: 45.14 -> 46.41Hurt
add the ReLU on the weighted combination of feature maps (ablation)Grad-CAM · ILSVRC-15 localizationlocalization performanceImproved
fuse Grad-CAM with Guided Backpropagation (Guided Grad-CAM)Guided Backpropagation · PASCAL VOC 2007 human studyhuman accuracy for class identification: 44.44% -> 61.23%Improved
combine Grad-CAM with DeconvolutionDeconvolution · PASCAL VOC 2007 human studyhuman accuracy for class identification: 53.33% -> 61.23%Improved
use Guided Grad-CAM instead of Guided Backpropagation to compare VGG and AlexNettrust evaluation · PASCAL VOC 2007 human studyrelative reliability score of VGG: 1.00 -> 1.27Improved
use Grad-CAM instead of Guided Backpropagation, c-MWP or CAM for faithfulness to occlusionVGG-16 · PASCAL VOC 2007 valrank correlation with occlusion map: 0.168 -> 0.254Improved
remove gender bias by adding male nurses and female doctors to training dataVGG-16 doctor vs nurse classifier · doctor vs nurse test setaccuracy: 82% -> 90%Improved
use Grad-CAM instead of Guided Backpropagation for localizing captionsneuraltalk2 captioning · DenseCap region boxesinside/outside activation ratio: 2.32 -> 3.27Improved
add Grad-CAM to Guided Backpropagation (Guided Grad-CAM) for caption localizationneuraltalk2 captioning · DenseCap region boxesinside/outside activation ratio: 3.27 -> 6.38Improved
use Grad-CAM instead of Guided Backpropagation for VQA faithfulnessVQA LSTM-CNN model · VQArank correlation with occlusion map: 0.42 -> 0.60Improved

Later papers that compared with this method

10 later accepted papers report GradCAM as a baseline in their results tables; 9 of them report a better number on most of the results they share with it.

2 more not listed.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy