SiamixFormer: a fully-transformer Siamese network with temporal Fusion for accurate building detection and change detection in bi-temporal remote sensing images
arXiv:2208.00657v2 Announce Type: cross Abstract: Building detection and change detection using remote sensing images can help urban and rescue planning. Moreover, they can be used for building damage assessment after natural disasters. Currently, most of the existing models for building detection use only one image (pre-disaster image) to detect buildings. This is based on the idea that post-disaster images reduce the model's performance because of presence of destroyed buildings. In this paper, we propose a siamese model, called SiamixFormer, which uses pre- and post-disaster images as input
The continuous advancements in AI, particularly in computer vision and transformer models, are enabling more sophisticated and accurate remote sensing applications for disaster assessment and urban planning.
This development improves post-disaster assessment capabilities, which is critical for humanitarian response, infrastructure resilience, and efficient resource allocation in a world facing increasing natural disasters.
The ability to accurately detect buildings and damage using bi-temporal images shifts disaster assessment from manual or single-image methods to more automated, precise, and comparative analyses.
- · Humanitarian organizations
- · Disaster relief agencies
- · Urban planners
- · Insurance companies
- · Manual damage assessment services
- · Organizations relying on outdated assessment methodologies
Faster and more accurate initial damage assessments post-disaster.
Improved efficiency in resource allocation for reconstruction and aid distribution.
Reduced economic losses and faster recovery times for affected regions due to superior intelligence.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI