arXiv:2605. 31162v1 Announce Type: cross Abstract: Unconditional diffusion models offer powerful generative priors, yet steering them toward aesthetically enhanced outputs remains largely unexplored.
Recent diffusion editors perform diverse instruction-based edits while conditioning on the source image at every denoising step. Yet persistent source-image conditioning can limit how fully an edit is executed and how natural the result appears, especially when the target scene diverges substantially from the input.
The paper introduces "overpainting," a localized, context-aware image editing technique that allows users to specify precise or loose editing regions via a trimap. The method adapts a pretrained diffusion model with joint attention and low‑rank adaptation, incorporating attention‑dropout to balance noise, source, and mask inputs. An automated pipeline generates training data by pairing images from language‑based editing models, curating them, and extracting trimaps, enabling the model to perform a wide range of editing tasks.
Diffusion Editing with Soft Mask: Pixel Level Redo of Image and Video with Adjustable Strength introduces SoftPaint, a zero‑shot sampling method that uses soft masks to provide continuous, pixel‑level control over edits in diffusion models. The approach employs a Langevin‑iteration sampler that respects per‑pixel mask strengths, enabling smooth edits from preserving to fully re‑synthesizing content across image and video backbones. SoftPaint is gradient‑free, memory‑efficient, and works universally with existing diffusion models.
arXiv:2510. 17136v2 Announce Type: replace Abstract: The generation of high-quality, diverse, and prompt-aligned images is a central goal in image-generating diffusion models.
SR-Edit is a new image editing framework that uses iterative self‑refinement to improve fidelity. At each step it extracts precise, self‑consistent region separations from the model’s predictions and then enforces preservation in non‑edit areas with correction updates that stay aligned with the original sampling dynamics. Experiments show that SR‑Edit delivers better preservation and overall image quality than existing editing techniques.
Despite remarkable progress in text-guided image editing, generative models frequently fail to preserve visual object consistency, defined as the preservation of a subject's key attributes throughout the editing process. We address this limitation through three contributions.
arXiv:2607. 19895v1 Announce Type: cross Abstract: Text-guided video editing with diffusion models is impractically slow, hindered by costly multi-step sampling and inversion.
arXiv:2606. 04299v1 Announce Type: cross Abstract: We consider the problem of generating images whose internal structure -- defined by the distribution of patches across multiple scales -- matches that of a single reference image.
Recent advances in diffusion models have shown impressive performance in controllable image generation and dense prediction tasks. However, existing approaches typically treat diffusion-based controllable generation and dense prediction as separate tasks, overlooking the potential benefits of jointly modeling the heterogeneous distributions.
arXiv:2606. 14125v1 Announce Type: cross Abstract: Inversion-based image editing offers flexible and training-free control but still struggles with inversion accuracy and the trade-off between editing fidelity and background preservation.
Unified image restoration (UIR) aims to recover high-quality (HQ) content from low-quality (LQ) images with different degradations using a single model. Most recent methods adapt large pretrained text-to-image (T2I) latent diffusion models for their strong capacity and generative priors.
The paper introduces an adaptive step schedule controller for text‑to‑image diffusion models, allowing the number of denoising steps to vary based on the complexity of the input prompt. By mixing step schedules of different sizes and monitoring error discrepancies at each timestep, the method switches schedules to maintain image quality while reducing inference time. Experiments on COCO and DiffusionDB demonstrate that this approach achieves faster generation without sacrificing visual fidelity.