Research map / Label-efficient and robust learning / Recognition and long-tail
Fixing Localization Errors to Improve Image Classification
Guolei Sun, Salman Khan, Wen Li, Hisham Cholakkal, Fahad Shahbaz Khan, Luc Van Gool
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 HNCmse loss | ResNet-56 CE · CIFAR-100 | top-1 accuracy: 72.40 -> 73.35 | Improved |
| add HNCmse loss | ResNet-110 CE · CIFAR-100 | top-1 accuracy: 73.79 -> 75.00 | Improved |
| add HNCmse loss | ResNet-101 CE · ImageNet | top-1 error: 23.2 -> 22.3 | Improved |
| add HNCkd loss | ResNet-152 CE · ImageNet | top-1 error: 22.9 -> 21.8 | Improved |
| add HNCmse loss | ResNet-50 CE · CUB-200-2011 | accuracy: 86.0 -> 87.1 | Improved |
| add HNCmse loss | ResNet-101 · MS-COCO | mAP: 74.9 -> 77.8 | Improved |
| add HNCmse loss | Attention Consistency (AC) · MS-COCO | mAP: 77.5 -> 78.5 | Improved |
| add HNCkd loss | ResNet-110 CE · CIFAR-100 FGSM eps 0.05 | accuracy: 14.55 -> 29.80 | Improved |
| add HNCmse loss | ResNet-110 CE · CIFAR-100 symmetric noise r 0.6 | accuracy: 22.48 -> 32.86 | Improved |
| add HNCmse loss | ResNet-110 CE · CIFAR-100 asymmetric noise r 0.4 | accuracy: 40.89 -> 46.51 | Improved |
| add HNC loss (ablation) | ResNeXt-29 c=8 CE · CIFAR-100 | accuracy: 81.77 -> 82.32 | Improved |
| add HNC loss (ablation) | DenseNet-BC CE · CIFAR-100 | accuracy: 77.32 -> 78.78 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Anchor Loss — Anchor Loss: Modulating Loss Scale Based on Prediction Difficulty (ICCV 2019) · better on 1 of 1 shared result
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
- Towards Learning Spatially Discriminative Feature Representations (ICCV 2021 · similarity 0.91)
- Keep CALM and Improve Visual Feature Attribution (ICCV 2021 · similarity 0.89)
- Finer-CAM: Spotting the Difference Reveals Finer Details for Visual Explanation (CVPR 2025 · similarity 0.89)
- DD-CAM: Minimal Sufficient Explanations for Vision Models Using Delta Debugging (NeurIPS 2026 · similarity 0.88)
- CAPE: CAM as a Probabilistic Ensemble for Enhanced DNN Interpretation (CVPR 2024 · similarity 0.88)
- MoRe: Class Patch Attention Needs Regularization for Weakly Supervised Semantic Segmentation (AAAI 2025 · similarity 0.87)
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.
