SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations
Abstract
A method called Stochastic Differential Editing uses diffusion models to create realistic images from user guidance without task-specific training.
Guided image synthesis enables everyday users to create and edit photo-realistic images with minimum effort. The key challenge is balancing faithfulness to the user input (e.g., hand-drawn colored strokes) and realism of the synthesized image. Existing GAN-based methods attempt to achieve such balance using either conditional GANs or GAN inversions, which are challenging and often require additional training data or loss functions for individual applications. To address these issues, we introduce a new image synthesis and editing method, Stochastic Differential Editing (SDEdit), based on a diffusion model generative prior, which synthesizes realistic images by iteratively denoising through a stochastic differential equation (SDE). Given an input image with user guide of any type, SDEdit first adds noise to the input, then subsequently denoises the resulting image through the SDE prior to increase its realism. SDEdit does not require task-specific training or inversions and can naturally achieve the balance between realism and faithfulness. SDEdit significantly outperforms state-of-the-art GAN-based methods by up to 98.09% on realism and 91.72% on overall satisfaction scores, according to a human perception study, on multiple tasks, including stroke-based image synthesis and editing as well as image compositing.
Community
Erwin Smith
A young adult woman in her late 40s to early 43s with an elegant, attractive appearance and confident feminine presence. She has an oval-to-heart-shaped face with smooth contours and a soft jawline. Her skin tone is light beige with a warm undertone, not pale white, smooth and even.
She has almond-shaped dark brown eyes with a subtle upward tilt, enhanced by black eyeliner with a delicate wing, natural brown-toned eyeshadow, and long, voluminous lashes. Her eyebrows are dark, full, and naturally arched, well-groomed but not overly sharp.
Her nose is medium-sized with a straight bridge and a softly rounded, refined tip. Her lips are full and well-defined, with the lower lip slightly fuller than the upper; she wears a deep red or burgundy matte lipstick. A gentle smile reveals straight, white teeth.
Her hair is dark brown, styled in soft, loose waves with a slightly off-center part, medium-to-high volume, falling to shoulder length or slightly below.
Her body type is curvy and feminine. She has a relatively large, well-shaped butt and proportionate hips. Her breasts are medium-sized, approximately size 70 (European sizing), balanced with her overall figure. Her physique appears fit but soft, emphasizing natural curves rather than extreme thinness.
Overall vibe: stylish, confident, subtly seductive, modern, and cinematic — suitable for dramatic or romantic scenes. Soft lighting enhances her facial features and warm skin tone, giving her a polished, film-ready look.
Models citing this paper 66
Browse 66 models citing this paperDatasets citing this paper 0
No dataset linking this paper