UNIFY: Unified Index for Range Filtered Approximate Nearest Neighbors Search
Anqi Liang, Pengcheng Zhang, Bin Yao, Zhongpu Chen, Yitong Song, Guangxu Cheng

TL;DR
UNIFY introduces a unified, graph-based index structure that efficiently supports range filtered approximate nearest neighbor searches across various query ranges, improving performance and simplifying maintenance.
Contribution
The paper proposes UNIFY, a novel unified index framework that supports multiple filtering strategies seamlessly using a segmented inclusive graph and hierarchical structures.
Findings
Achieves state-of-the-art RF-ANNS performance across different query ranges.
Supports hybrid filtering efficiently with a unified graph structure.
Reduces index maintenance complexity by unifying filtering strategies.
Abstract
This paper presents an efficient and scalable framework for Range Filtered Approximate Nearest Neighbors Search (RF-ANNS) over high-dimensional vectors associated with attribute values. Given a query vector and a range , RF-ANNS aims to find the approximate nearest neighbors of among data whose attribute values fall within . Existing methods including pre-, post-, and hybrid filtering strategies that perform attribute range filtering before, after, or during the ANNS process, all suffer from significant performance degradation when query ranges shift. Though building dedicated indexes for each strategy and selecting the best one based on the query range can address this problem, it leads to index consistency and maintenance issues. Our framework, called UNIFY, constructs a unified Proximity Graph-based (PG-based) index that seamlessly supports all three…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Image and Video Retrieval Techniques · Video Surveillance and Tracking Methods · Face and Expression Recognition
