AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Composite Quantization

Jingdong Wang, Ting Zhang

TPAMI 2018 · Architectures and efficiency › Quantization · approach: Post-training weight quantization · 30 citations (OpenAlex)

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 changedWhereMeasuredEffect
relax orthogonality to near-orthogonality constraint (ablation)Near-orthogonal composite quantization · 1M SIFT 64 bitsRecall@10: 0.643 -> 0.715Improved
use near-orthogonal composite quantization instead of Cartesian k-meansNear-orthogonal composite quantization · 1M SIFTRecall@10 (T=1, 64 bits): 63.83 -> 71.59Improved
use NOCQ instead of product quantizationNear-orthogonal composite quantization · 1M SIFTRecall@10 (T=1, 64 bits): 60.45 -> 71.59Improved
use NOCQ instead of CKMNear-orthogonal composite quantization · 1M CNN 64 bitsRecall@10 (T=1): 30.28 -> 36.07Improved
use NOCQ instead of CKMNear-orthogonal composite quantization · 1B SIFT 64 bitsRecall@100 (T=1): 64.57 -> 70.12Improved
discard the third inter-dictionary term at search time (CQ-d) instead of NOCQ constraint (ablation)Near-orthogonal composite quantization · 1M SIFT 64 bitsRecall@10: 0.715 -> 0.388Hurt
add near-orthogonality constraint in place of storing third term as extra 4 bytes (ablation)Near-orthogonal composite quantization · 1M SIFT 64 bitsRecall@10: 0.769 -> 0.715Mixed
add near-orthogonality constraint (ablation)Near-orthogonal composite quantization · 1M SIFT 64 bitsquantization error: 1.54 -> 1.65Hurt
learn epsilon instead of fixing it at zero (ablation)Near-orthogonal composite quantization · 1M SIFTrecallNo clear effect
learn epsilon instead of fixing it at zero (ablation)Near-orthogonal composite quantization · 1B SIFTrecallImproved
add a global translation offset (ablation)Near-orthogonal composite quantization · 1M SIFT 64 bitsrecallNo clear effect
quantize augmented vector (QAV) instead of near-orthogonal CQ on data vectors (ablation)Near-orthogonal composite quantization · 1M SIFT 64 bitsRecall@10: 0.715 -> 0.355Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy