From Image Hashing to Scene Change Detection

2026-05-12

AI Summary

This work addresses the limitation of traditional image hashing methods, which support only global comparison and cannot localize specific regions where scene changes occur. To overcome this, we propose HashSCD—the first unsupervised image hashing framework capable of spatially localized change detection. Our approach employs spatially aligned patch-wise hash codes and integrates contrastive learning at both patch and global levels during training. A novel XOR-like operation is designed to aggregate local hash codes in Hamming space, enabling simultaneous global change detection and pixel-level localization through a single forward pass. Experimental results demonstrate that HashSCD achieves comparable accuracy to state-of-the-art methods while significantly reducing computational overhead and storage requirements.

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.

Problem

• image hashing
• scene change detection
• spatial localization
• change detection
• Hamming space

Innovation

• patch-wise hashing
• scene change detection
• Hamming space
• unsupervised contrastive learning
• change localization