{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/changen2-multi-temporal-remote-sensing","title":"Changen2: Multi-Temporal Remote Sensing Generative Change Foundation Model","arxiv_id":"2406.17998","date":"2024-06-26","proceeding":null,"authors":["Zhuo Zheng","Stefano Ermon","Dongjun Kim","Liangpei Zhang","Yanfei Zhong"],"abstract":"Our understanding of the temporal dynamics of the Earth's surface has been advanced by deep vision models, which often require lots of labeled multi-temporal images for training. However, collecting, preprocessing, and annotating multi-temporal remote sensing images at scale is non-trivial since it is expensive and knowledge-intensive. In this paper, we present change data generators based on generative models, which are cheap and automatic, alleviating these data problems. Our main idea is to simulate a stochastic change process over time. We describe the stochastic change process as a probabilistic graphical model (GPCM), which factorizes the complex simulation problem into two more tractable sub-problems, i.e., change event simulation and semantic change synthesis. To solve these two problems, we present Changen2, a GPCM with a resolution-scalable diffusion transformer which can generate time series of images and their semantic and change labels from labeled or unlabeled single-temporal images. Changen2 is a generative change foundation model that can be trained at scale via self-supervision, and can produce change supervisory signals from unlabeled single-temporal images. Unlike existing foundation models, Changen2 synthesizes change data to train task-specific foundation models for change detection. The resulting model possesses inherent zero-shot change detection capabilities and excellent transferability. Experiments suggest Changen2 has superior spatiotemporal scalability, e.g., Changen2 model trained on 256$^2$ pixel single-temporal images can yield time series of any length and resolutions of 1,024$^2$ pixels. Changen2 pre-trained models exhibit superior zero-shot performance (narrowing the performance gap to 3% on LEVIR-CD and approximately 10% on both S2Looking and SECOND, compared to fully supervised counterparts) and transferability across multiple types of change tasks.","url_abs":"https://arxiv.org/abs/2406.17998v1","url_pdf":"https://arxiv.org/pdf/2406.17998v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"changen2-multi-temporal-remote-sensing","repo_url":"https://github.com/Z-Zheng/pytorch-change-models","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"change-detection","task_name":"Change Detection"},{"task_slug":"time-series-1","task_name":"Time Series"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[{"slug":"changen2-s1-15k","name":"Changen2-S1-15k","full_name":"Changen2-S1-15k"},{"slug":"changen2-s9-27k","name":"Changen2-S9-27k","full_name":"Changen2-S9-27k"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2406.17998","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}