Research map / Label-efficient and robust learning / Recognition and long-tail
Fast Online Hashing with Multi-Label Projection
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 the query pool to restrict query search and code updates to candidate neighbors | FOH · CIFAR-10 and FLICKR-25K | query time: 8.46 -> 2.18; 5.27 -> 1.33; 6.87 -> 2.45; 3.05 -> 2.85 | Helped |
| add label projection loss (ablation) | FOH · CIFAR-10, 32-bits | mAP: 0.717 -> 0.734 | Helped |
| add label projection loss (ablation) | FOH · FLICKR-25K, 32-bits | mAP: 0.583 -> 0.604 | Helped |
| use the proposed multi-label similarity matrix instead of the traditional one (ablation) | FOH · FLICKR-25K, 32-bits | mAP: 0.554 -> 0.598 | Helped |
| use the proposed multi-label similarity matrix instead of the traditional one (ablation) | FOH · FLICKR-25K, 48-bits | mAP: 0.555 -> 0.605 | Helped |
| add the query pool (ablation) | FOH · CIFAR-10 and FLICKR-25K | mAP: 0.747 vs 0.734; 0.595 vs 0.598 | No clear effect |
| reduce central point update frequency from 1-b to 5-b (ablation) | FOH · CIFAR-10, 32-bits | mAP: 0.73403 -> 0.72529 | Hurt |
| reduce central point update frequency from 1-b to 5-b (ablation) | FOH · FLICKR-25K, 32-bits | mAP: 0.60448 -> 0.59135 | Hurt |
Closest papers by meaning
- Online Hashing with Efficient Updating of Binary Codes (AAAI 2020 · similarity 0.92)
- Towards Optimal Discrete Online Hashing with Balanced Similarity (AAAI 2019 · similarity 0.88)
- Deep Graph Online Hashing for Multi-Label Image Retrieval (AAAI 2025 · similarity 0.88)
- Supervised Hashing via Uncorrelated Component Analysis (AAAI 2016 · similarity 0.88)
- Generalized Debiased Semi-Supervised Hashing for Large-Scale Image Retrieval (AAAI 2025 · similarity 0.86)
- Pairwise-Label-Based Deep Incremental Hashing with Simultaneous Code Expansion (AAAI 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.