SENSE: Satellite-based ENergy Synthesis for Sustainable Environment

SENSE (Satellite-based ENergy Synthesis for Sustainable Environment) is a unified generative Urban Building Energy Modeling (UBEM) framework. It jointly synthesizes realistic urban satellite imagery along with aligned building energy consumption and height maps using a controllable diffusion model.

• Paper: SENSE: Satellite-based ENergy Synthesis for Sustainable Environment
• GitHub: GenAI4Urban-Energy
• Dataset: skl24/MUSE

Description

By conditioning on road networks and urban density metrics, SENSE leverages knowledge learned by large vision models to generate urban building energy consumption and height annotations in the latent space. Experiments across multiple cities (New York City, Boston, Lyon, Busan) demonstrate that SENSE achieves high visual fidelity and strong physical consistency, satisfying ASHRAE standard metrics.

Citation

@article{sun2024sense,
  title={SENSE: Satellite-based ENergy Synthesis for Sustainable Environment},
  author={Sun, Kailai and He, Mingyi and Huang, Heye and Rong, Can and Prakash, Alok and Guo, Baoshen and Wang, Shenhao and Zhao, Jinhua},
  journal={arXiv preprint arXiv:2605.18101},
  year={2024}
}