Research map / Detection and segmentation / Open-set and few-shot detection
Picking Deep Filter Responses for Fine-Grained Image Recognition
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 automatically learned part detection to whole-image FC-CNN features (ablation) | VGG-M recognition pipeline · CUB-200-2011 | accuracy: 66.97% -> 76.74% | Helped |
| Add automatically learned part detection to whole-image FC-CNN features (ablation) | AlexNet recognition pipeline · Stanford Dogs | accuracy: 59.67% -> 65.07% | Helped |
| Use part detection with FV-CNN instead of whole-image FV-CNN (ablation) | VGG-M recognition pipeline · CUB-200-2011 | accuracy: 58.71% -> 73.83% | Helped |
| Use part detection with FV-CNN instead of whole-image FV-CNN (ablation) | AlexNet recognition pipeline · Stanford Dogs | accuracy: 60.52% -> 63.11% | Helped |
| Combine whole-image FC-CNN and FV-CNN features (ablation) | VGG-M recognition pipeline · CUB-200-2011 | accuracy: 66.97% -> 71.03% | Helped |
| Combine whole-image FC-CNN and FV-CNN features (ablation) | AlexNet recognition pipeline · Stanford Dogs | accuracy: 59.67% -> 63.75% | Helped |
| Replace FV-CNN with spatially weighted SWFV-CNN (ablation) | PD+FV-CNN · CUB-200-2011 | accuracy: 73.83% -> 77.26% | Helped |
| Replace FV-CNN with spatially weighted SWFV-CNN (ablation) | PD+FV-CNN · Stanford Dogs | accuracy: 63.11% -> 66.25% | Helped |
| Combine FC-CNN with spatially weighted SWFV-CNN (ablation) | VGG-M recognition pipeline with part detection · CUB-200-2011 | accuracy: 76.74% -> 80.26% | Helped |
| Combine FC-CNN with spatially weighted SWFV-CNN (ablation) | AlexNet recognition pipeline with part detection · Stanford Dogs | accuracy: 65.07% -> 71.96% | Helped |
| Use the deeper VGG-VD model instead of VGG-M (ablation) | PD+FC+SWFV-CNN · CUB-200-2011 | accuracy: 80.26% -> 84.54% | Helped |
| Use the full automatic method rather than NAC | CUB-200-2011 · CUB-200-2011 | accuracy: 81.01% -> 84.54% | Helped |
Closest papers by meaning
- Fine-Grained Recognition as HSnet Search for Informative Image Parts (CVPR 2017 · similarity 0.87)
- Fine-Grained Image Classification via Combining Vision and Language (CVPR 2017 · similarity 0.87)
- Multi-scale Activation, Refinement, and Aggregation: Exploring Diverse Cues for Fine-Grained Bird Recognition (AAAI 2025 · similarity 0.85)
- Democratizing Fine-grained Visual Recognition with Large Language Models (ICLR 2024 · similarity 0.85)
- Prompt-CAM: Making Vision Transformers Interpretable for Fine-Grained Analysis (CVPR 2025 · similarity 0.84)
- Fine-grained Classes and How to Find Them (ICML 2024 · similarity 0.84)
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.