mlsanityData to 7 Oct 2026

Opti-CAM: Optimizing saliency maps for interpretability

arXiv 2301.07002 ↗First posted 17 Jan 2023cs.CVcited by 25

Not on the map. Why?

Abstract (arXiv)

Methods based on class activation maps (CAM) provide a simple mechanism to interpret predictions of convolutional neural networks by using linear combinations of feature maps as saliency maps. By contrast, masking-based methods optimize a saliency map directly in the image space or learn it by training another network on additional data. In this work we introduce Opti-CAM, combining ideas from CAM-based and masking-based approaches. Our saliency map is a linear combination of feature maps, where weights are optimized per image such that the logit of the masked image for a given class is maximized. We also fix a fundamental flaw in two of the most common evaluation metrics of attribution methods. On several datasets, Opti-CAM largely outperforms other CAM-based approaches according to the most relevant classification metrics. We provide empirical evidence supporting that localization and classifier interpretability are not necessarily aligned.

Key facts. Cited by 25 papers in the map (data through 7 Oct 2026). Builds on 53 of its 68 references that are in the map. First posted on arXiv 17 Jan 2023.

Builds on

53 of 68 references.

References · 53

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  5. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  6. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  7. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  8. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  9. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  10. 2020Training data-efficient image transformers & distillation through attentioncited by 4,560
  11. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  12. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
38 more
  1. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  2. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  3. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  4. 2018Sanity Checks for Saliency Mapscited by 1,149
  5. 2017SmoothGrad: removing noise by adding noisecited by 1,097
  6. 2014Striving for Simplicity: The All Convolutional Netcited by 983
  7. 2018A Survey Of Methods For Explaining Black Box Modelscited by 970
  8. 2017Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networkscited by 936
  9. 2018RISE: Randomized Input Sampling for Explanation of Black-box Modelscited by 825
  10. 2020Quantifying Attention Flow in Transformerscited by 759
  11. 2020Transformer Interpretability Beyond Attention Visualizationcited by 620
  12. 2017Interpretable Explanations of Black Boxes by Meaningful Perturbationcited by 611
  13. 2019Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networkscited by 575
  14. 2015Evaluating the visualization of what a Deep Neural Network has learnedcited by 416
  15. 2017Methods for Interpreting and Understanding Deep Neural Networkscited by 342
  16. 2015Understanding Neural Networks Through Deep Visualizationcited by 332
  17. 2016Top-down Neural Attention by Excitation Backpropcited by 318
  18. 2019On the (In)fidelity and Sensitivity for Explanationscited by 309
  19. 2019Understanding Deep Networks via Extremal Perturbations and Smooth Maskscited by 257
  20. 2017Real Time Image Saliency for Black Box Classifierscited by 246
  21. 2019Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relationscited by 245
  22. 2016Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentationcited by 198
  23. 2020Eigen-CAM: Class Activation Map using Principal Componentscited by 188
  24. 2015Weakly Supervised Deep Detection Networkscited by 183
  25. 2022Quantus: An Explainable AI Toolkit for Responsible Evaluation of Neural Network Explanations and Beyondcited by 171
  26. 2020Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNscited by 151
  27. 2018Explaining Image Classifiers by Counterfactual Generationcited by 144
  28. 2021Benchmarking and Survey of Explanation Methods for Black Box Modelscited by 136
  29. 2020The elephant in the interpretability room: Why use attention as explanation when we have saliency methods?cited by 122
  30. 2021Interpretable Deep Learning: Interpretation, Interpretability, Trustworthiness, and Beyondcited by 116
  31. 2020Restricting the Flow: Information Bottlenecks for Attributioncited by 110
  32. 2020Evaluating Weakly Supervised Object Localization Methods Rightcited by 104
  33. 2020There and Back Again: Revisiting Backpropagation Saliency Methodscited by 60
  34. 2018Not Using the Car to See the Sidewalk: Quantifying and Controlling the Effects of Context in Classification and Segmentationcited by 49
  35. 2020SS-CAM: Smoothed Score-CAM for Sharper Visual Feature Localizationcited by 46
  36. 2018Local Explanation Methods for Deep Neural Networks Lack Sensitivity to Parameter Valuescited by 40
  37. 2022Metrics for saliency map evaluation of deep learning explanation methodscited by 40
  38. 2022Towards Better Understanding Attribution Methodscited by 28

50 most cited of 53 All 53 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 4; 2024: 9; 2025: 10; 2026: 2.

25 papers.

Sort:

  1. 2025JAFAR: Jack up Any Feature at Any Resolutioncited by 21
  2. 2025DocVXQA: Context-Aware Visual Explanations for Document Question Answeringcited by 5
  3. 2023UPDExplainer: an Interpretable Transformer-based Framework for Urban Physical Disorder Detection Using Street View Imagerycited by 4
  4. 2024CAM-Based Methods Can See through Wallscited by 4
  5. 2025What Does an Audio Deepfake Detector Focus on? A Study in the Time Domaincited by 4
  6. 2024Listenable Maps for Zero-Shot Audio Classifierscited by 3
  7. 2023SPADE: Sparsity-Guided Debugging for Deep Neural Networkscited by 2
  8. 2023On the stability, correctness and plausibility of visual explanation methods based on feature importancecited by 2
  9. 2024CA-Stream: Attention-based pooling for interpretable image recognitioncited by 2
  10. 2024PEPL: Precision-Enhanced Pseudo-Labeling for Fine-Grained Image Classification in Semi-Supervised Learningcited by 2
  11. 2024The Overfocusing Bias of Convolutional Neural Networks: A Saliency-Guided Regularization Approachcited by 2
  12. 2025Generating visual explanations from deep networks using implicit neural representationscited by 1
13 more
  1. 2025There is More to Attention: Statistical Filtering Enhances Explanations in Vision Transformerscited by 1
  2. 2025Back to the Baseline: Examining Baseline Effects on Explainability Metricscited by 1
  3. 2023Explanation-based Training with Differentiable Insertion/Deletion Metric-aware Regularizerscited by 0
  4. 2024Foveated Retinotopy Improves Classification and Localization in Convolutional Neural Networkscited by 0
  5. 2024A Learning Paradigm for Interpretable Gradientscited by 0
  6. 2024CNN Explainability with Multivector Tucker Saliency Maps for Self-Supervised Modelscited by 0
  7. 2024Advancing Stroke Risk Prediction Using a Multi-modal Foundation Modelcited by 0
  8. 2025P-TAME: Explain Any Image Classifier with Trained Perturbationscited by 0
  9. 2025CF-CAM: Cluster Filter Class Activation Mapping for Reliable Gradient-Based Interpretabilitycited by 0
  10. 2025WebXAII: an open-source web framework to study human-XAI interactioncited by 0
  11. 2025CaberNet: Causal Representation Learning for Cross-Domain HVAC Energy Predictioncited by 0
  12. 2026AGRI-Fidelity: Evaluating the Reliability of Listenable Explanations for Poultry Disease Detectioncited by 0
  13. 2026Signal or Noise? Auditing Rotation-Induced Saliency Drift in Medical and Aerial Imagingcited by 0