GeoDiT: Point-Conditioned Diffusion Transformer for Satellite Image Synthesis

2026-03-02

AI Summary

This work proposes a point-conditioned diffusion Transformer framework for satellite image generation, addressing the limitations of existing methods that rely on pixel-level maps with limited semantics and high annotation costs. By leveraging sparse spatial coordinate points and their associated textual descriptions as guidance, the approach enables semantically rich and annotation-efficient controllable synthesis. Key innovations include a point-query-driven adaptive local attention mechanism, geographic coordinate embeddings, and a multimodal text-spatial alignment strategy. Experimental results demonstrate that the proposed method outperforms state-of-the-art models in remote sensing image generation across multiple metrics, significantly improving both generation quality and controllability.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Generation
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

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.

Problem

• satellite image synthesis
• point-based control
• conditional generation
• remote sensing
• semantic control

Innovation

• point-conditioned diffusion
• satellite image synthesis
• adaptive local attention
• text-to-image generation
• geospatial conditioning