Research map / Label-efficient and robust learning / Domain adaptation and generalization
Text-Phase Synergy Network with Dual Priors for Unsupervised Cross-Domain Image Retrieval
Jing Yang, Hui Xue, Shipeng Zhu, Pengfei Fang
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 domain prompts (DPG) as text-prior supervision (ablation) | TPSNet baseline (ViT-B) · Office-Home (avg 12 scenarios) | P@1: 65.18 -> 77.09 | Improved |
| add domain prompts (DPG) as text-prior supervision (ablation) | TPSNet baseline (ViT-B) · DomainNet (avg 12 scenarios) | P@50: 70.94 -> 77.63 | Improved |
| add phase prior (PPFE) on top of domain prompts (ablation) | TPSNet (ViT-B) · Office-Home (avg 12 scenarios) | P@1: 77.09 -> 81.34 | Improved |
| add text-prior feature extraction (TPFE) on top of domain prompts (ablation) | TPSNet (ViT-B) · Office-Home (avg 12 scenarios) | P@1: 77.09 -> 81.63 | Improved |
| combine TPFE and PPFE (dual priors) instead of one (ablation) | TPSNet (ViT-B) · DomainNet (avg 12 scenarios) | P@50: 82.05 -> 85.33 | Improved |
| replace CLIP image encoder with MoCov2 init in TPSNet (ablation) | TPSNet (ResNet-50) · Office-Home (avg 12 scenarios) | P@1: 61.80 -> 56.71 | Hurt |
| replace CLIP image encoder with DINO init in TPSNet (ablation) | TPSNet (ViT-B) · Office-Home (avg 12 scenarios) | P@1: 84.53 -> 67.62 | Hurt |
| TPSNet vs prior SOTA with ViT-B | UCDIR · Office-Home (avg 12 scenarios) | P@1: 64.71 -> 84.53 | Improved |
| TPSNet vs ShieldIR with ResNet-50 on abstract Painting->Quickdraw | UCDIR · DomainNet Painting->Quickdraw | P@50: 65.33 -> 67.21 | Improved |
Closest papers by meaning
- Semantic Feature Learning for Universal Unsupervised Cross-Domain Retrieval (NeurIPS 2024 · similarity 0.90)
- Unsupervised Cross-Domain Image Retrieval via Prototypical Optimal Transport (AAAI 2024 · similarity 0.89)
- Retrieval Across Any Domains via Large-scale Pre-trained Model (ICML 2024 · similarity 0.89)
- RoDA: Robust Domain Alignment for Cross-Domain Retrieval Against Label Noise (AAAI 2025 · similarity 0.87)
- DiDA: Disambiguated Domain Alignment for Cross-Domain Retrieval with Partial Labels (AAAI 2024 · similarity 0.86)
- ProS: Prompting-to-simulate Generalized knowledge for Universal Cross-Domain Retrieval (CVPR 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.
