Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Doodle to Object: Practical Zero-Shot Sketch-Based 3D Shape Retrieval
Read the paperCheck 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 |
|---|---|---|---|
| use PCL instead of CGN | ZS-SBSR retrieval · SHREC’13 | mAP: 38.6 -> 48.0 | Helped |
| use PCL instead of CGN | ZS-SBSR retrieval · SHREC’14 | mAP: 35.6 -> 43.4 | Helped |
| use PCL instead of CGN | ZS-SBSR retrieval · D2O | mAP: 25.0 -> 29.9 | Helped |
| add cross-domain feature alignment (ablation) | PCL · SHREC’13 ZS-SBSR | mAP: 23.2 -> 48.0 | Helped |
| add cross-domain feature alignment (ablation) | PCL · SHREC’14 ZS-SBSR | mAP: 24.1 -> 43.4 | Helped |
| add cross-domain feature alignment (ablation) | PCL · D2O ZS-SBSR | mAP: 16.5 -> 29.9 | Helped |
| combine prototype and contrastive learning rather than use dynamic clustering (ablation) | cross-domain aligned retrieval · SHREC’13 ZS-SBSR | mAP: 34.7 -> 48.0 | Helped |
| combine prototype and contrastive learning rather than use fixed anchor points (ablation) | cross-domain aligned retrieval · SHREC’14 ZS-SBSR | mAP: 30.3 -> 43.4 | Helped |
| use 80% stroke accumulation with the RNN rather than only the final sketch image (ablation) | D2O retrieval · SBSR | mAP: 51.3 -> 50.8 | No clear effect |
| use 100% stroke accumulation with the RNN rather than only the final sketch image (ablation) | D2O retrieval · ZS-SBSR | mAP: 29.9 -> 30.4 | No clear effect |
Closest papers by meaning
- Doodle to Search: Practical Zero-Shot Sketch-Based Image Retrieval (CVPR 2019 · similarity 0.90)
- CLIP for All Things Zero-Shot Sketch-Based Image Retrieval, Fine-Grained or Not (CVPR 2023 · similarity 0.88)
- Sketch3T: Test-Time Training for Zero-Shot SBIR (CVPR 2022 · similarity 0.88)
- Learning Barycentric Representations of 3D Shapes for Sketch-Based 3D Shape Retrieval (CVPR 2017 · similarity 0.87)
- How to Handle Sketch-Abstraction in Sketch-Based Image Retrieval? (CVPR 2024 · similarity 0.87)
- Asymmetric Mutual Alignment for Unsupervised Zero-Shot Sketch-Based Image Retrieval (AAAI 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.