Towards Remote Sensing Change Detection with Neural Memory

2026-02-11

AI Summary

This work addresses the challenge of simultaneously achieving computational efficiency and effective modeling of long-range spatiotemporal dependencies in remote sensing change detection. To this end, we propose ChangeTitans, a novel framework that introduces the Titans architecture to this domain for the first time. Our approach features a vision backbone, VTitans, built upon neural memory and segmented local attention, along with a hierarchical VTitans-Adapter and a dual-stream temporal cross-attention module (TS-CBAM) to suppress pseudo-changes and capture long-range inter-temporal dependencies. Evaluated on four benchmarks including LEVIR-CD, the method achieves state-of-the-art performance, attaining an IoU of 84.36% and an F1-score of 91.52% on LEVIR-CD, demonstrating both high accuracy and computational efficiency.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution

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.

Problem

• remote sensing change detection
• long-range dependencies
• computational efficiency
• spatiotemporal relationships

Innovation

• neural memory
• segmented local attention
• VTitans
• TS-CBAM
• change detection