Changes in Gaza: DINOv3-Powered Multi-Class Change Detection for Damage Assessment in Conflict Zones

TL;DR

This paper introduces a novel multi-scale cross-attention change detection method using a pre-trained DINOv3 model for accurate damage assessment in conflict zones, demonstrated on a new Gaza-change dataset and classical datasets.

Contribution

The paper presents a new multi-scale cross-attention Siamese network with DINOv3 backbone for fine-grained semantic change detection in conflict zones, along with a new Gaza-change dataset.

Findings

Effective detection of subtle semantic changes in conflict zones.

Superior performance on Gaza-Change, SECOND, and Landsat-SCD datasets.

High accuracy and efficiency in damage assessment applications.

Abstract

Accurately and swiftly assessing damage from conflicts is crucial for humanitarian aid and regional stability. In conflict zones, damaged zones often share similar architectural styles, with damage typically covering small areas and exhibiting blurred boundaries. These characteristics lead to limited data, annotation difficulties, and significant recognition challenges, including high intra-class similarity and ambiguous semantic changes. To address these issues, we introduce a pre-trained DINOv3 model and propose a multi-scale cross-attention difference siamese network (MC-DiSNet). The powerful visual representation capability of the DINOv3 backbone enables robust and rich feature extraction from bi-temporal remote sensing images. The multi-scale cross-attention mechanism allows for precise localization of subtle semantic changes, while the difference siamese structure enhances inter-class feature discrimination, enabling fine-grained semantic change detection. Furthermore, a simple yet powerful lightweight decoder is designed to generate clear detection maps while maintaining high efficiency. We also release a new Gaza-change dataset containing high-resolution satellite image pairs from 2023-2024 with pixel-level semantic change annotations. It is worth emphasizing that our annotations only include semantic pixels of changed areas. We evaluated our method on the Gaza-Change and two classical datasets: the SECOND and Landsat-SCD datasets. Experimental results demonstrate that our proposed approach effectively addresses the MCD task, and its outstanding performance paves the way for practical applications in rapid damage assessment across conflict zones.