Image Hashing via Cross-View Code Alignment in the Age of Foundation Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Moummad, Ilyass, Zaher, Kawtar, Goëau, Hervé, Joly, Alexis
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915914370252800
author Moummad, Ilyass
Zaher, Kawtar
Goëau, Hervé
Joly, Alexis
author_facet Moummad, Ilyass
Zaher, Kawtar
Goëau, Hervé
Joly, Alexis
contents Efficient large-scale retrieval requires representations that are both compact and discriminative. Foundation models provide powerful visual and multimodal embeddings, but nearest neighbor search in these high-dimensional spaces is computationally expensive. Hashing offers an efficient alternative by enabling fast Hamming distance search with binary codes, yet existing approaches often rely on complex pipelines, multi-term objectives, designs specialized for a single learning paradigm, and long training times. We introduce CroVCA (Cross-View Code Alignment), a simple and unified principle for learning binary codes that remain consistent across semantically aligned views. A single binary cross-entropy loss enforces alignment, while coding-rate maximization serves as an anti-collapse regularizer to promote balanced and diverse codes. To implement this, we design HashCoder, a lightweight MLP hashing network with a final batch normalization layer to enforce balanced codes. HashCoder can be used as a probing head on frozen embeddings or to adapt encoders efficiently via LoRA fine-tuning. Across benchmarks, CroVCA achieves state-of-the-art results in just 5 training epochs. At 16 bits, it performs particularly well; for instance, unsupervised hashing on COCO completes in under 2 minutes and supervised hashing on ImageNet100 in about 3 minutes on a single GPU. These results highlight CroVCA's efficiency, adaptability, and broad applicability.
format Preprint
id arxiv_https___arxiv_org_abs_2510_27584
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Image Hashing via Cross-View Code Alignment in the Age of Foundation Models
Moummad, Ilyass
Zaher, Kawtar
Goëau, Hervé
Joly, Alexis
Computer Vision and Pattern Recognition
Information Retrieval
Machine Learning
Efficient large-scale retrieval requires representations that are both compact and discriminative. Foundation models provide powerful visual and multimodal embeddings, but nearest neighbor search in these high-dimensional spaces is computationally expensive. Hashing offers an efficient alternative by enabling fast Hamming distance search with binary codes, yet existing approaches often rely on complex pipelines, multi-term objectives, designs specialized for a single learning paradigm, and long training times. We introduce CroVCA (Cross-View Code Alignment), a simple and unified principle for learning binary codes that remain consistent across semantically aligned views. A single binary cross-entropy loss enforces alignment, while coding-rate maximization serves as an anti-collapse regularizer to promote balanced and diverse codes. To implement this, we design HashCoder, a lightweight MLP hashing network with a final batch normalization layer to enforce balanced codes. HashCoder can be used as a probing head on frozen embeddings or to adapt encoders efficiently via LoRA fine-tuning. Across benchmarks, CroVCA achieves state-of-the-art results in just 5 training epochs. At 16 bits, it performs particularly well; for instance, unsupervised hashing on COCO completes in under 2 minutes and supervised hashing on ImageNet100 in about 3 minutes on a single GPU. These results highlight CroVCA's efficiency, adaptability, and broad applicability.
title Image Hashing via Cross-View Code Alignment in the Age of Foundation Models
topic Computer Vision and Pattern Recognition
Information Retrieval
Machine Learning
url https://arxiv.org/abs/2510.27584