Research map / Learning theory and algorithms / Prediction and decision losses
Ordinal Constrained Binary Code Learning for Nearest Neighbor Search
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 OCH instead of the strongest listed baseline | unsupervised hashing · LabelMe, 32-bit codes | mAP: 0.3059 -> 0.3140 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · LabelMe, 64-bit codes | mAP: 0.3849 -> 0.3947 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · LabelMe, 128-bit codes | mAP: 0.4477 -> 0.4620 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · LabelMe, 32-bit codes | Pre@100: 0.5012 -> 0.5072 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · LabelMe, 64-bit codes | Pre@100: 0.5920 -> 0.6028 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · LabelMe, 128-bit codes | Pre@100: 0.6589 -> 0.6713 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · Tiny100K, 32-bit codes | mAP: 0.2054 -> 0.2297 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · Tiny100K, 64-bit codes | mAP: 0.2637 -> 0.2919 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · Tiny100K, 128-bit codes | mAP: 0.3193 -> 0.3379 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · Tiny100K, 32-bit codes | Pre@100: 0.4444 -> 0.4805 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · Tiny100K, 64-bit codes | Pre@100: 0.5599 -> 0.5785 | Helped |
| Use OCH instead of the strongest listed baseline | unsupervised hashing · Tiny100K, 128-bit codes | Pre@100: 0.6755 -> 0.6314 | Hurt |
Closest papers by meaning
- Towards Optimal Binary Code Learning via Ordinal Embedding (AAAI 2016 · similarity 0.92)
- Discretely Coding Semantic Rank Orders for Supervised Image Hashing (CVPR 2017 · similarity 0.88)
- Hashing as Tie-Aware Learning to Rank (CVPR 2018 · similarity 0.87)
- K-Nearest Neighbors Hashing (CVPR 2019 · similarity 0.87)
- Learning to Hash Robustly, Guaranteed (ICML 2022 · similarity 0.86)
- Learning Attribute-Aware Hash Codes for Fine-Grained Image Retrieval via Query Optimization (ICML 2025 · similarity 0.85)
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.