Research map / Learning theory and algorithms / Prediction and decision losses
Efficient Querying from Weighted Binary Codes
Zhenyu Weng, Yuesheng Zhu
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 |
|---|---|---|---|
| Use weighted-distance-ordered bucket search with multi-index tables and exact candidate merging instead of linear scanning | Weighted binary-code querying on Places205, GIST1M, and SIFT1B · Places205, GIST1M, and SIFT1B | precision@K: 25.29 vs 25.29; 13.20 vs 13.20; 3.25 vs 3.25 (MIWQ vs Linear Scan, 32-bit 1-NN) | No clear effect |
| Use weighted-distance-ordered bucket search with multi-index tables and exact candidate merging instead of linear scanning | Weighted binary-code querying on SIFT1B · SIFT1B (32-bit, 1-NN) | speed-up factor: 1.0 -> 3639.0 | Improved |
| Use exact weighted-distance candidate selection rather than MIH's Hamming-based candidate filtering | MIWQ vs MIH · SIFT1B (64-bit, 100-NN) | precision (%): 10.52 -> 14.13 | Improved |
| Use MIWQ rather than MIH | SIFT1B (64-bit, 100-NN) · SIFT1B (64-bit, 100-NN) | speed-up factor: 1899.9 -> 1356.0 | Hurt |
| Use the proposed table-merging algorithm rather than PQTable's merging algorithm | PQTable with max-heap merging · SIFT1B (32-bit, 1-NN) | query time (ms): 50.44 -> 6.15 | Improved |
| Use MIWQ's bucket-finding process rather than PQTable's | Exact weighted binary-code search · SIFT1B (32-bit, 1-NN) | query time (ms): 50.44 -> 6.15 | Improved |
Closest papers by meaning
- Online Nearest Neighbor Search Using Hamming Weight Trees (TPAMI 2019 · similarity 0.86)
- Improved Hamming Distance Search Using Variable Length Substrings (CVPR 2016 · similarity 0.84)
- Simple and Efficient Weighted Minwise Hashing (NeurIPS 2016 · similarity 0.83)
- Structured Learning of Binary Codes with Column Generation for Optimizing Ranking Measures (IJCV 2017 · similarity 0.83)
- Re-randomized Densification for One Permutation Hashing and Bin-wise Consistent Weighted Sampling (NeurIPS 2019 · similarity 0.82)
- Online Hashing with Efficient Updating of Binary Codes (AAAI 2020 · similarity 0.82)
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.
