From Image Hashing to Scene Change Detection

TL;DR

HashSCD introduces a patch-wise hashing framework for efficient and localized scene change detection, enabling global and spatial change analysis without heavy computation.

Contribution

The paper proposes HashSCD, a novel unsupervised patch-wise hashing method that localizes scene changes efficiently in the Hamming space.

Findings

HashSCD achieves competitive accuracy with state-of-the-art methods.

It significantly reduces computational cost and storage requirements.

The model enables both global and localized change detection in an unsupervised manner.

Abstract

Image hashing provides compact representations for efficient storage and retrieval but is inherently limited to global comparison and cannot reason about where changes occur. This limitation prevents hashing from being directly applicable to scene change detection, where spatial localization is essential. In this work, we revisit hashing from a scene change detection perspective and propose HashSCD, a patch-wise hashing framework that enables both efficient global change detection and localized change identification. HashSCD encodes spatially aligned patches into compact hash codes and aggregates them through an XOR-like operation, allowing change detection and localization to be performed directly in the Hamming space without repeated inference on previous images. The model is trained in an unsupervised manner using contrastive learning at both patch and global levels. Experiments demonstrate that HashSCD achieves competitive performance compared to state-of-the-art unsupervised hashing and scene change detection methods, while significantly reducing computational cost and storage requirements.