Research map / Label-efficient and robust learning / Recognition and long-tail
Hadamard Matrix Guided Online Hashing
Mingbao Lin, Rongrong Ji, Hong Liu, Xiaoshuai Sun, Chen Shen, Qi Tian
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 |
|---|---|---|---|
| replace linear regression on tanh-relaxed codes (HCOH) with perceptron bit classifiers, LSH-aligned Hadamard targets and model ensembling | HCOH · CIFAR-10 | mAP: average increase of 4.926% | Improved |
| replace linear regression with perceptron classifiers on Hadamard targets (HMOH vs HCOH) at low code length | HCOH · Places205 (8-bit) | mAP@1000: increase of 108.163% | Improved |
| replace linear regression with perceptron classifiers on Hadamard targets (HMOH vs HCOH) at 16 bits | HCOH · Places205 (16-bit) | mAP@1000: 34.104% | Improved |
| HMOH vs HCOH, averaged over bit lengths | HCOH · MNIST | mAP: 15.595% | Improved |
| HMOH vs best baseline (MIHash or BSODH), averaged over bit lengths | BSODH / MIHash · CIFAR-10 | mAP: average percentage of 7.478% | Improved |
| increase code length from 64 to 128 bits | HMOH · CIFAR-10 | Precision@H2: 0.710 -> 0.734 | Improved |
| increase code length from 64 to 128 bits | HMOH · Places205 | Precision@H2: 0.223 -> 0.137 | Hurt |
| increase code length from 64 to 128 bits | HMOH · NUS-WIDE | Precision@H2: 0.474 -> 0.468 | No clear effect |
| use more training data (2K to 20K samples) | HMOH · MNIST (48-bit) | mAP: 0.780 -> 0.845 | Improved |
| add kernelization (RBF anchors) (ablation) | HMOH · CIFAR-10 and NUS-WIDE | mAP | Hurt |
| add ensembling of models learned in every round (ablation) | HMOH · benchmarks | mAP | Improved |
| more training data for low bit codes (8, 16, 32 bits) | HMOH · NUS-WIDE | mAP | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MIHash — MIHash: Online Hashing With Mutual Information (ICCV 2017) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Fast Class-wise Updating for Online Hashing (TPAMI 2020 · similarity 0.89)
- Towards Optimal Discrete Online Hashing with Balanced Similarity (AAAI 2019 · similarity 0.87)
- Hashing with Binary Matrix Pursuit (ECCV 2018 · similarity 0.87)
- Deep Graph Online Hashing for Multi-Label Image Retrieval (AAAI 2025 · similarity 0.86)
- MIHash: Online Hashing With Mutual Information (ICCV 2017 · similarity 0.86)
- Fast Supervised Discrete Hashing (TPAMI 2017 · 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.
