SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations. SDEdit enables image editing and synthesis by leveraging stochastic differential equations, supporting both conditional (image-to-image) and unconditional (text-guided) diffusion. The pipeline allows for flexible editing while preserving content and structure of the source image. [Meng et al., 2022]
NULL-text Inversion for Editing Real Images Using Guided Diffusion Models. NULL-text Inversion enables editing of real images using guided diffusion models through text inversion techniques. [Mokady et al., 2023]
Taming Rectified Flow for Inversion and Editing. Taming Rectified Flow enables image inversion and editing using rectified flow models. [Wang et al., 2025]
Tuning-Based Editing
References
Meng, Chenlin; He, Yutong; Song, Yang; Song, Jiaming; Wu, Jiajun; Zhu, Jun-Yan; Ermon, Stefano (2022). SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations. International Conference on Learning Representations (ICLR).
Hertz, Amir; Mokady, Ron; Tenenbaum, Jay; Aberman, Kfir; Pritch, Yael; Cohen-or, Daniel (2022). Prompt-to-Prompt Image Editing with Cross-Attention Control. The Eleventh International Conference on Learning Representations.
Couairon, Guillaume; Verbeek, Jakob; Schwenk, Holger; Cord, Matthieu (2023). DiffEdit: Diffusion-based Semantic Image Editing with Mask Guidance. The Eleventh International Conference on Learning Representations.
Mokady, Ron; Hertz, Amir; Aberman, Kfir; Pritch, Yael; Cohen-Or, Daniel (2023). NULL-text Inversion for Editing Real Images Using Guided Diffusion Models. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).