{"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/lanpaint-training-free-diffusion-inpainting","title":"Lanpaint: Training-Free Diffusion Inpainting with Exact and Fast Conditional Inference","arxiv_id":"2502.03491","date":"2025-02-05","proceeding":null,"authors":["Candi Zheng","Yuan Lan","Yang Wang"],"abstract":"Diffusion models generate high-quality images but often lack efficient and universally applicable inpainting capabilities, particularly in community-trained models. We introduce LanPaint, a training-free method tailored for widely adopted ODE-based samplers, which leverages Langevin dynamics to perform exact conditional inference, enabling precise and visually coherent inpainting. LanPaint addresses two key challenges in Langevin-based inpainting: (1) the risk of local likelihood maxima trapping and (2) slow convergence. By proposing a guided score function and a fast-converging Langevin framework, LanPaint achieves high-fidelity results in very few iterations. Experiments demonstrate that LanPaint outperforms existing training-free inpainting techniques, outperforming in challenging tasks such as outpainting with Stable Diffusion.","url_abs":"https://arxiv.org/abs/2502.03491v1","url_pdf":"https://arxiv.org/pdf/2502.03491v1.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":"lanpaint-training-free-diffusion-inpainting","repo_url":"https://github.com/scraed/LanPaint","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"},{"method_slug":"pixel-prediction","method_name":"Inpainting"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2502.03491","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}