Research map / Vision-language and multimodal / CLIP and zero-shot
Learning to Parameterize Visual Attributes for Open-set Fine-grained Retrieval
Shijie Wang, Jianlong Chang, Haojie Li, Zhihui Wang, Wanli Ouyang, Qi Tian
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 original constraint Lc to La (ablation) | VAPNet · CUB-200-2011 | Recall@1: 69.5 -> 71.4 | Improved |
| add original constraint Lc to La (ablation) | VAPNet · Stanford Cars | Recall@1: 89.3 -> 91.2 | Improved |
| use parameterization constraint Lc-hat instead of Lc (without La) (ablation) | VAPNet · CUB-200-2011 | Recall@1: 71.4 -> 74.1 | Improved |
| add auxiliary loss La to Lc-hat (ablation) | VAPNet · CUB-200-2011 | Recall@1: 74.1 -> 76.2 | Improved |
| add auxiliary loss La to Lc-hat (ablation) | VAPNet · Stanford Cars | Recall@1: 92.7 -> 94.8 | Improved |
| increase local views M from 1 to 4 (ablation) | VAPNet · CUB-200-2011 | Recall@1: 73.6 -> 76.2 | Improved |
| increase local views M from 4 to 16 (ablation) | VAPNet · CUB-200-2011 | Recall@1: 76.2 -> 76.2 | No clear effect |
| attribute dimension k from 32 to 256 (ablation) | VAPNet · CUB-200-2011 | Recall@1: 74.7 -> 76.2 | Improved |
| attribute dimension k from 256 to 512 (ablation) | VAPNet · CUB-200-2011 | Recall@1: 76.2 -> 76.0 | Hurt |
| increase updating ratio alpha from 0.2 to 0.8 (ablation) | VAPNet · CUB-200-2011 | Recall@1: 76.2 -> 72.6 | Hurt |
| updating ratio alpha 0.1 vs 0.2 (ablation) | VAPNet · CUB-200-2011 | Recall@1: 75.9 -> 76.2 | No clear effect |
| balance weight lambda from 1 to 10 (ablation) | VAPNet · CUB-200-2011 | Recall@1: 73.8 -> 76.2 | Improved |
Closest papers by meaning
- Language-driven Fine-grained Retrieval (CVPR 2026 · similarity 0.89)
- Learning Deep Representations of Fine-Grained Visual Descriptions (CVPR 2016 · similarity 0.87)
- OW-VAP: Visual Attribute Parsing for Open World Object Detection (ICML 2025 · similarity 0.87)
- Democratizing Fine-grained Visual Recognition with Large Language Models (ICLR 2024 · similarity 0.85)
- Highlighting What Matters: Promptable Embeddings for Attribute-Focused Image Retrieval (NeurIPS 2025 · similarity 0.85)
- GUIDED: Granular Understanding via Identification, Detection, and Discrimination for Fine-Grained Open-Vocabulary Object Detection (NeurIPS 2025 · similarity 0.85)
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.
