GeoDiT: Point-Conditioned Diffusion Transformer for Satellite Image Synthesis
2026-03-02 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors present GeoDiT, a new model that creates satellite images from text descriptions using simple point markers instead of detailed maps. This method makes it easier and faster to guide the image generation process with meaningful spatial and descriptive clues. They also developed a special attention technique to better focus on these points during image creation. Tests show GeoDiT generates better satellite images than previous methods. The authors also explored different ways to represent satellite images and locations for the best results.
diffusion modeltransformertext-to-image generationsatellite imagerypoint-based conditioninglocal attentionsemantic controlgeolocationremote sensingimage generation
Authors
Srikumar Sastry, Dan Cher, Brian Wei, Aayush Dhakal, Subash Khanal, Dev Gupta, Nathan Jacobs
Abstract
We introduce GeoDiT, a diffusion transformer designed for text-to-satellite image generation with point-based control. Existing controlled satellite image generative models often require pixel-level maps that are time-consuming to acquire, yet semantically limited. To address this limitation, we introduce a novel point-based conditioning framework that controls the generation process through the spatial location of the points and the textual description associated with each point, providing semantically rich control signals. This approach enables flexible, annotation-friendly, and computationally simple inference for satellite image generation. To this end, we introduce an adaptive local attention mechanism that effectively regularizes the attention scores based on the input point queries. We systematically evaluate various domain-specific design choices for training GeoDiT, including the selection of satellite image representation for alignment and geolocation representation for conditioning. Our experiments demonstrate that GeoDiT achieves impressive generation performance, surpassing the state-of-the-art remote sensing generative models.