Research map / Detection and segmentation / CNN features
Multi-Attention Multi-Class Constraint for Fine-grained Image Recognition
Ming Sun, Yuchen Yuan, Feng Zhou, Errui Ding
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 OSME module without MAMC (ablation) | ResNet-50 · CUB-200-2011 | top-1 accuracy (%): 81.7 -> 84.9 | Improved |
| use OSME instead of two independent FC layers under MAMC (ablation) | ResNet-50 + MAMC · CUB-200-2011 | top-1 accuracy (%): 85.7 -> 86.2 | Improved |
| add first MAMC constraint on top of OSME (ablation) | ResNet-50 + OSME · CUB-200-2011 | top-1 accuracy (%): 84.9 -> 85.4 | Improved |
| add the remaining two MAMC constraints (ablation) | ResNet-50 + OSME + MAMC · CUB-200-2011 | top-1 accuracy (%): 85.4 -> 86.2 | Improved |
| increase attention regions from P=2 to P=3 (ablation) | ResNet-50 + OSME + MAMC · CUB-200-2011 | top-1 accuracy (%): 86.2 -> 86.3 | No clear effect |
| use ResNet-101 instead of ResNet-50 backbone (ablation) | OSME + MAMC · CUB-200-2011 | top-1 accuracy (%): 86.2 -> 86.5 | Improved |
| full method versus ResNet-50 baseline | ResNet-50 · CUB-200-2011 | top-1 accuracy (%): 81.7 -> 86.2 | Improved |
| full method versus RACNN, multi-stage | RACNN · Dogs-in-the-Wild | top-1 accuracy (%): 76.5 -> 77.9 | Improved |
| full method versus RACNN on Stanford Dogs | RACNN · Stanford Dogs | top-1 accuracy (%): 87.3 -> 85.2 | Hurt |
| ResNet-101 OSME+MAMC vs ResNet-101 baseline on Stanford Cars | ResNet-101 · Stanford Cars | top-1 accuracy (%): 91.9 -> 93.0 | Improved |
Closest papers by meaning
- Look Closer to See Better: Recurrent Attention Convolutional Neural Network for Fine-Grained Image Recognition (CVPR 2017 · similarity 0.91)
- Learning Multi-Attention Convolutional Neural Network for Fine-Grained Image Recognition (ICCV 2017 · similarity 0.91)
- Looking for the Devil in the Details: Learning Trilinear Attention Sampling Network for Fine-Grained Image Recognition (CVPR 2019 · similarity 0.90)
- Dual Attention Networks for Few-Shot Fine-Grained Recognition (AAAI 2022 · similarity 0.90)
- Improving Fine-Grained Visual Recognition in Low Data Regimes via Self-Boosting Attention Mechanism (ECCV 2022 · similarity 0.89)
- Part-Stacked CNN for Fine-Grained Visual Categorization (CVPR 2016 · similarity 0.88)
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.
