Research map / Vision-language and multimodal / CLIP and zero-shot
Semantically Tied Paired Cycle Consistency for Zero-Shot Sketch-Based Image Retrieval
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 |
|---|---|---|---|
| Use SEM-PCYC instead of the strongest listed zero-shot SBIR baseline | Zero-shot SBIR retrieval · Sketchy (Extended) | mAP@all: 0.258 -> 0.349 | Helped |
| Use SEM-PCYC instead of the strongest listed zero-shot SBIR baseline | Zero-shot SBIR retrieval · TU-Berlin (Extended) | mAP@all: 0.223 -> 0.297 | Helped |
| Add cycle consistency to adversarial training (ablation) | SEM-PCYC · Sketchy (Extended) | mAP@all: 0.128 -> 0.147 | Helped |
| Add cycle consistency to adversarial training (ablation) | SEM-PCYC · TU-Berlin (Extended) | mAP@all: 0.109 -> 0.131 | Helped |
| Add classification loss to adversarial training (ablation) | SEM-PCYC · Sketchy (Extended) | mAP@all: 0.128 -> 0.140 | Helped |
| Add classification loss to adversarial training (ablation) | SEM-PCYC · TU-Berlin (Extended) | mAP@all: 0.109 -> 0.127 | Helped |
| Remove side-information selection and use original-dimensional side information (ablation) | SEM-PCYC · Sketchy (Extended) | mAP@all: 0.349 -> 0.382 | Helped |
| Remove side-information selection and use original-dimensional side information (ablation) | SEM-PCYC · TU-Berlin (Extended) | mAP@all: 0.297 -> 0.299 | Helped |
| Remove the auto-encoder regularizer (ablation) | SEM-PCYC · Sketchy (Extended) | mAP@all: 0.349 -> 0.323 | Hurt |
| Remove the auto-encoder regularizer (ablation) | SEM-PCYC · TU-Berlin (Extended) | mAP@all: 0.297 -> 0.273 | Hurt |
| Combine Word2Vec with Jiang-Conrath hierarchical similarity rather than Word2Vec alone (ablation) | SEM-PCYC side information · Sketchy (Extended) | mAP@all: 0.330 -> 0.349 | Helped |
| Combine Word2Vec with path similarity rather than Word2Vec alone (ablation) | SEM-PCYC side information · TU-Berlin (Extended) | mAP@all: 0.232 -> 0.297 | Helped |
Closest papers by meaning
- Zero-Shot Sketch-Based Image Retrieval via Graph Convolution Network (AAAI 2020 · similarity 0.93)
- Asymmetric Mutual Alignment for Unsupervised Zero-Shot Sketch-Based Image Retrieval (AAAI 2024 · similarity 0.92)
- ARNet: Self-Supervised FG-SBIR with Unified Sample Feature Alignment and Multi-Scale Token Recycling (AAAI 2025 · similarity 0.89)
- You'll Never Walk Alone: A Sketch and Text Duet for Fine-Grained Image Retrieval (CVPR 2024 · similarity 0.88)
- Text-to-Image Diffusion Models are Great Sketch-Photo Matchmakers (CVPR 2024 · similarity 0.88)
- Open Vocabulary Semantic Scene Sketch Understanding (CVPR 2024 · 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.