Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

large-scale image retrieval

Large-scale image retrieval is the process of efficiently searching, identifying, and retrieving visually or semantically similar images from massive databases containing millions or billions of images in response to a query image. Unlike retrieval systems operating on smaller datasets, large-scale retrieval must overcome significant computational and memory constraints to achieve real-time search speeds without sacrificing accuracy. To accomplish this, systems extract visual features from images and compress them into compact representations using techniques such as binary hashing, vector quantization, and dimensionality reduction. These compact representations enable scalable storage and rapid approximate nearest neighbor search, allowing the system to quickly match queries against massive repositories.

1 item

Iterative Quantization: A Procrustean Approach to Learning Binary Codes for Large-Scale Image Retrieval

Iterative Quantization: A Procrustean Approach to Learning Binary Codes for Large-Scale Image Retrieval

Yunchao Gong, Svetlana Lazebnik, Albert Gordo, Florent Perronnin

OrganizationsUniversitat Autònoma de BarcelonaUniversity of Illinois Urbana-ChampaignUniversity of North Carolina at Chapel HillXerox

Why you should read this

Proposes an alternating minimization method, Iterative Quantization, that finds an optimal orthogonal rotation to align dimensionally-reduced data with the vertices of a binary hypercube, significantly improving retrieval accuracy and scalability for large-scale image search.

This paper addresses the problem of learning similarity-preserving binary codes for efficient similarity search in large-scale image collections. We formulate this problem in terms of finding a rotation of zero-centered data so as to minimize the quantization error of mapping this data to the vertices of a zero-centered binary hypercube, and propose a simple and efficient alternating minimization algorithm to accomplish this task. This algorithm, dubbed iterative quantization (ITQ), has connections to multi-class spectral clustering and to the orthogonal Procrustes problem, and it can be used both with unsupervised data embeddings such as PCA and supervised embeddings such as canonical correlation analysis (CCA). The resulting binary codes significantly outperform several other state-of-the-art methods. We also show that further performance improvements can result from transforming the data with a nonlinear kernel mapping prior to PCA or CCA. Finally, we demonstrate an application of ITQ to learning binary attributes or “classemes” on the ImageNet dataset.

Added

2026-09-24