Imagen de Portada

Reference-Vector Removed Product Quantization for Approximate Nearest Neighbor Search

Guardado en:
Publicado en:Applied Sciences vol. 15, no. 21 (2025), p. 11540-11561
Autor principal: Wang, Yang
Otros Autores: Xu, Ce, Wu, Xueyi
Publicado:
MDPI AG
Materias:
Decomposition
Accuracy
Codes
Algorithms
Norms
Data compression
Image retrieval
Acceso en línea:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Resumen:This paper proposes a decorrelation scheme based on product quantization, termed Reference-Vector Removed Product Quantization (RvRPQ), for approximate nearest neighbor (ANN) search. The core idea is to capture the redundancy among database vectors by representing them with compactly encoded reference-vectors, which are then subtracted from the original vectors to yield residual vectors. We provide a theoretical derivation for obtaining the optimal reference-vectors. This preprocessing step significantly improves the quantization accuracy of the subsequent product quantization applied to the residuals. To maintain low online computational complexity and control memory overhead, we apply vector quantization to the reference-vectors and allocate only a small number of additional bits to store their indices. Experimental results show that RvRPQ substantially outperforms state-of-the-art ANN methods in terms of retrieval accuracy, while preserving high search efficiency.
ISSN:2076-3417
DOI:10.3390/app152111540
Fuente:Publicly Available Content Database