Research map / Architectures and efficiency / CNN design
Winsor-CAM: Human-Tunable Visual Explanations From Deep Networks via Layer-Wise Winsorization
Casey Wall, L. Wang, Rodrigue Rizk, KC Santosh
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add layer-wise Winsorized aggregation over all conv layers instead of final-layer Grad-CAM | Grad-CAM (DenseNet121) · PASCAL VOC 2012 subset (correct) | IoU: 0.390 -> 0.468 | Improved |
| add Winsorization instead of naive mean aggregation of layer maps | naive layer averaging (DenseNet121) · PASCAL VOC 2012 subset (correct) | IoU: 0.428 -> 0.468 | Improved |
| add Winsorization instead of final-layer Grad-CAM | Grad-CAM (DenseNet121) · PASCAL VOC 2012 subset (correct) | CoM distance: 0.074 -> 0.059 | Improved |
| add Winsorization instead of final-layer Grad-CAM | Grad-CAM (DenseNet121) · PASCAL VOC 2012 subset (correct) | Deletion AUC: 0.242 -> 0.197 | Improved |
| add Winsorization instead of final-layer Grad-CAM | Grad-CAM (VGG16) · PASCAL VOC 2012 subset (correct) | Insertion AUC: 0.900 -> 0.873 | Hurt |
| add Winsorization instead of final-layer Grad-CAM | Grad-CAM (EfficientNet-B0) · PASCAL VOC 2012 subset (correct) | IoU: 0.398 -> 0.391 | Hurt |
| use bilinear interpolation instead of nearest-neighbor (ablation) | Winsor-CAM (DenseNet121, mean) · PASCAL VOC 2012 subset (correct) | IoU: 0.456 -> 0.468 | Improved |
| use mean aggregation instead of max aggregation (ablation) | Winsor-CAM (DenseNet121, bilinear) · PASCAL VOC 2012 subset (correct) | IoU: 0.453 -> 0.468 | Improved |
| use max aggregation instead of mean aggregation (ablation) | Winsor-CAM (VGG16, bilinear) · PASCAL VOC 2012 subset (correct) | Insertion AUC: 0.873 -> 0.905 | Improved |
| increase fixed percentile p from 0 to 50 (ablation) | Winsor-CAM (DenseNet121) · PASCAL VOC 2012 subset (correct) | IoU: 0.442 -> 0.445 | No clear effect |
| increase fixed percentile p from 0 to 90 (ablation) | Winsor-CAM (DenseNet121) · PASCAL VOC 2012 subset (correct) | Deletion AUC: 0.208 -> 0.215 | Hurt |
| tune p per image (best over p) instead of one fixed p (ablation) | Winsor-CAM (DenseNet121) · PASCAL VOC 2012 subset (correct) | IoU: 0.445 -> 0.468 | Improved |
Closest papers by meaning
- DD-CAM: Minimal Sufficient Explanations for Vision Models Using Delta Debugging (NeurIPS 2026 · similarity 0.87)
- Empowering CAM-Based Methods with Capability to Generate Fine-Grained and High-Faithfulness Explanations (AAAI 2024 · similarity 0.85)
- Finer-CAM: Spotting the Difference Reveals Finer Details for Visual Explanation (CVPR 2025 · similarity 0.84)
- CAPE: CAM as a Probabilistic Ensemble for Enhanced DNN Interpretation (CVPR 2024 · similarity 0.84)
- Attention Guided CAM: Visual Explanations of Vision Transformer Guided by Self-Attention (AAAI 2024 · similarity 0.84)
- DiffCAM: Data-Driven Saliency Maps by Capturing Feature Differences (CVPR 2025 · similarity 0.83)
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.
