Image Retrieval
Image Retrieval is a fundamental and long-standing computer vision task that involves finding images similar to a given query from a large database. It is often considered a form of fine-grained, instance-level classification. The task is integral to image recognition alongside classification and cross-modal retrieval. By leveraging visual similarity and other criteria, image retrieval enables users to efficiently discover relevant images, making it a crucial tool in applications such as search and recommendation.
Extending CLIP for Category-to-image Retrieval in E-commerce
( Image credit: DELF )
Papers
Showing 1–10 of 2239 papers
All datasetsROxford (Hard)ROxford (Medium)RParis (Hard)RParis (Medium)CREPE (Compositional REPresentation Evaluation)Fashion IQFlickr30K 1K testCIRRSOPFlickr30k-CNOxf5kFlickr30k
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Unicom+ViT-L@336px | R@1 | 91.2 | — | Unverified |
| 2 | ROADMAP (DeiT-B) | R@1 | 86 | — | Unverified |
| 3 | CGD (SG/GS) | R@1 | 84.2 | — | Unverified |
| 4 | ROADMAP (ResNet-50) | R@1 | 83.1 | — | Unverified |
| 5 | ProxyNCA++ | R@1 | 81.4 | — | Unverified |
| 6 | PNP Loss | R@1 | 81.1 | — | Unverified |
| 7 | Cross-Batch Memory | R@1 | 80.6 | — | Unverified |
| 8 | Smooth-AP | R@1 | 80.1 | — | Unverified |
| 9 | NormSoftmax2048 (ResNet-50) | R@1 | 79.5 | — | Unverified |
| 10 | EPSHN512 | R@1 | 78.3 | — | Unverified |