Research map / Label-efficient and robust learning / Recognition and long-tail
Pairwise Confusion for Fine-Grained Visual Classification
Abhimanyu Dubey, Otkrist Gupta, Pei Guo, Ramesh Raskar, Ryan Farrell, Nikhil Naik
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 Pairwise Confusion | DenseNet-161 · CUB-200-2011 | top-1 accuracy: 84.21 -> 86.87 | Improved |
| add Pairwise Confusion | ResNet-50 · NABirds | top-1 accuracy: 63.55 -> 68.15 | Improved |
| add Pairwise Confusion | Bilinear CNN · six FGVC datasets | top-1 accuracy: average improvement of 1.7% | Improved |
| add Pairwise Confusion | GoogLeNet · ImageNet-Dogs (from scratch) | top-1 accuracy: 62.35 -> 64.17 | Improved |
| add Pairwise Confusion | GoogLeNet · ImageNet-Random (from scratch) | top-1 accuracy: 71.85 -> 72.09 | Improved |
| add Pairwise Confusion | ResNet-50 · CIFAR-10 (from scratch) | top-1 accuracy: 93.17 -> 93.46 | Improved |
| add Pairwise Confusion | DenseNet-161 · CIFAR-10 (from scratch) | top-1 accuracy: 95.15 -> 95.08 | No clear effect |
| add Pairwise Confusion | GoogLeNet · CIFAR-100 (from scratch) | top-1 accuracy: 73.35 -> 76.02 | Improved |
| add Pairwise Confusion | ResNet-50 · ImageNet-1K full (from scratch) | top-1 accuracy: 76.15 -> 76.28 | No clear effect |
| add Pairwise Confusion | GoogLeNet · CUB-200-2011 Grad-CAM localization | mean IoU: 0.29 -> 0.35 | Improved |
| add Pairwise Confusion | VGG-16 · CUB-200-2011 Grad-CAM localization | mean IoU: 0.31 -> 0.34 | Improved |
| replace Label Smoothing Regularization with Pairwise Confusion (ablation) | ResNet-50 · CUB-200-2011 | top-1 accuracy: 78.20 -> 80.21 | Improved |
Closest papers by meaning
- An Erudite Fine-Grained Visual Classification Model (CVPR 2023 · similarity 0.91)
- Fine-Grained Visual Classification via Progressive Multi-Granularity Training of Jigsaw Patches (ECCV 2020 · similarity 0.90)
- Maximum-Entropy Fine Grained Classification (NeurIPS 2018 · similarity 0.89)
- Interweaving Insights: High-Order Feature Interaction for Fine-Grained Visual Recognition (IJCV 2024 · similarity 0.89)
- Advancing Fine-Grained Classification by Structure and Subject Preserving Augmentation (NeurIPS 2024 · similarity 0.86)
- Novel Class Discovery for Ultra-Fine-Grained Visual Categorization (CVPR 2024 · similarity 0.86)
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.
