Research map / Label-efficient and robust learning / Domain adaptation and generalization
Feature Representation Learning for Unsupervised Cross-Domain Image Retrieval
Conghui Hu, Gim Hee Lee
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 cluster-wise contrastive loss L_CW to instance-wise loss (ablation) | L_IW baseline · DomainNet average | P@50 (%): 38.14 -> 41.57 | Improved |
| add cluster-wise contrastive loss L_CW to instance-wise loss (ablation) | L_IW baseline · DomainNet average | P@200 (%): 29.42 -> 32.31 | Improved |
| add self-entropy loss on clustering probabilities (ablation) | v2 (L_IW + L_CW) · DomainNet average | P@50 (%): 41.57 -> 44.61 | Improved |
| add self-entropy loss on clustering probabilities (ablation) | v2 (L_IW + L_CW) · DomainNet average | P@200 (%): 32.31 -> 36.01 | Improved |
| add distance-of-distance (DD) loss (ablation) | v3 (without DD loss) · DomainNet average | P@50 (%): 44.61 -> 47.09 | Improved |
| add distance-of-distance (DD) loss (ablation) | v3 (without DD loss) · DomainNet average | P@200 (%): 36.01 -> 39.09 | Improved |
| add DD loss on small domain-gap pair (ablation) | v3 · DomainNet Real->Infograph | P@50 (%): 57.08 -> 57.74 | Improved |
| add DD loss on large domain-gap pair (ablation) | v3 · DomainNet Real->Quickdraw | P@50 (%): 33.19 -> 41.90 | Improved |
| add DD loss for Sketch->Infograph (ablation) | v2 (L_IW + L_CW) · DomainNet Sketch->Infograph | P@50 (%): 43.81 -> 43.66 | Hurt |
| use proposed method instead of PCS on Office-Home | PCS · Office-Home average | P@1 (%): 38.07 -> 40.67 | Improved |
| use proposed method instead of PCS on DomainNet | PCS · DomainNet average | P@50 (%): 41.23 -> 47.09 | Improved |
Closest papers by meaning
- Semantic Feature Learning for Universal Unsupervised Cross-Domain Retrieval (NeurIPS 2024 · similarity 0.92)
- Unsupervised Cross-Domain Image Retrieval via Prototypical Optimal Transport (AAAI 2024 · similarity 0.91)
- Retrieval Across Any Domains via Large-scale Pre-trained Model (ICML 2024 · similarity 0.88)
- DiDA: Disambiguated Domain Alignment for Cross-Domain Retrieval with Partial Labels (AAAI 2024 · similarity 0.88)
- RoDA: Robust Domain Alignment for Cross-Domain Retrieval Against Label Noise (AAAI 2025 · similarity 0.88)
- Effective Comparative Prototype Hashing for Unsupervised Domain Adaptation (AAAI 2024 · similarity 0.87)
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.
