Towards Remote Sensing Change Detection with Neural Memory

TL;DR

This paper introduces ChangeTitans, a novel neural network framework using Titans-based vision backbone with neural memory and hierarchical modules for improved remote sensing change detection, achieving state-of-the-art accuracy and efficiency.

Contribution

It presents VTitans, the first Titans-based vision backbone with neural memory and local attention, along with a hierarchical adapter and a cross-temporal fusion module for enhanced change detection.

Findings

Achieves 84.36% IoU and 91.52% F1-score on LEVIR-CD.

Outperforms existing methods on four benchmark datasets.

Maintains computational efficiency while capturing long-range dependencies.

Abstract

Remote sensing change detection is essential for environmental monitoring, urban planning, and related applications. However, current methods often struggle to capture long-range dependencies while maintaining computational efficiency. Although Transformers can effectively model global context, their quadratic complexity poses scalability challenges, and existing linear attention approaches frequently fail to capture intricate spatiotemporal relationships. Drawing inspiration from the recent success of Titans in language tasks, we present ChangeTitans, the Titans-based framework for remote sensing change detection. Specifically, we propose VTitans, the first Titans-based vision backbone that integrates neural memory with segmented local attention, thereby capturing long-range dependencies while mitigating computational overhead. Next, we present a hierarchical VTitans-Adapter to refine multi-scale features across different network layers. Finally, we introduce TS-CBAM, a two-stream fusion module leveraging cross-temporal attention to suppress pseudo-changes and enhance detection accuracy. Experimental evaluations on four benchmark datasets (LEVIR-CD, WHU-CD, LEVIR-CD+, and SYSU-CD) demonstrate that ChangeTitans achieves state-of-the-art results, attaining \textbf{84.36\%} IoU and \textbf{91.52\%} F1-score on LEVIR-CD, while remaining computationally competitive.