{"ID":2822866,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2601.01487","arxiv_id":"2601.01487","title":"DeepInv: A Novel Self-supervised Learning Approach for Fast and Accurate Diffusion Inversion","abstract":"Diffusion inversion is a task of recovering the noise of an image in a diffusion model, which is vital for controllable diffusion image editing. At present, diffusion inversion still remains a challenging task due to the lack of viable supervision signals. Thus, most existing methods resort to approximation-based solutions, which however are often at the cost of performance or efficiency. To remedy these shortcomings, we propose a novel self-supervised diffusion inversion approach in this paper, termed Deep Inversion (DeepInv). Instead of requiring ground-truth noise annotations, we introduce a self-supervised objective as well as a data augmentation strategy to generate high-quality pseudo noises from real images without manual intervention. Based on these two innovative designs, DeepInv is also equipped with an iterative and multi-scale training regime to train a parameterized inversion solver, thereby achieving the fast and accurate image-to-noise mapping. To the best of our knowledge, this is the first attempt of presenting a trainable solver to predict inversion noise step by step. The extensive experiments show that our DeepInv can achieve much better performance and inference speed than the compared methods, e.g., +40.435% SSIM than EasyInv and +9887.5% speed than ReNoise on COCO dataset. Moreover, our careful designs of trainable solvers can also provide insights to the community. Codes and model parameters will be released in https://github.com/potato-kitty/DeepInv.","short_abstract":"Diffusion inversion is a task of recovering the noise of an image in a diffusion model, which is vital for controllable diffusion image editing. At present, diffusion inversion still remains a challenging task due to the lack of viable supervision signals. Thus, most existing methods resort to approximation-based solut...","url_abs":"https://arxiv.org/abs/2601.01487","url_pdf":"https://arxiv.org/pdf/2601.01487v1","authors":"[\"Ziyue Zhang\",\"Luxi Lin\",\"Xiaolin Hu\",\"Chao Chang\",\"HuaiXi Wang\",\"Yiyi Zhou\",\"Rongrong Ji\"]","published":"2026-01-04T11:27:26Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Diffusion Model\"]","has_code":false,"code_links":[{"ID":605459,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2822866,"paper_url":"https://arxiv.org/abs/2601.01487","paper_title":"DeepInv: A Novel Self-supervised Learning Approach for Fast and Accurate Diffusion Inversion","repo_url":"https://github.com/potato-kitty/DeepInv","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}