TL;DR本文研究了一种用于近似最近邻搜索的紧凑编码方法,介绍了一种复合量化框架,其中包括使用不同字典中选定的若干元素的组合来准确近似 D 维向量,以此表示数据向量,实现准确搜索,从而实现距离计算成本从 O(D)降至 O(M)的近正交复合量化方法。
Abstract
This paper studies the compact coding approach to approximate nearest
neighbor search. We introduce a composite quantization framework. It uses the
composition of several ($M$) elements, each of which is selected