Stable Flow: Vital Layers for Training-Free Image Editing

• URL: http://arxiv.org/abs/2411.14430v1
• Date: Thu, 21 Nov 2024 18:59:51 GMT
• Title: Stable Flow: Vital Layers for Training-Free Image Editing
• Authors: Omri Avrahami, Or Patashnik, Ohad Fried, Egor Nemchinov, Kfir Aberman, Dani Lischinski, Daniel Cohen-Or,
• Abstract summary: Diffusion models have revolutionized the field of content synthesis and editing. Recent models have replaced the traditional UNet architecture with the Diffusion Transformer (DiT) We propose an automatic method to identify "vital layers" within DiT, crucial for image formation. Next, to enable real-image editing, we introduce an improved image inversion method for flow models.
• Score: 74.52248787189302
• License:
• Abstract: Diffusion models have revolutionized the field of content synthesis and editing. Recent models have replaced the traditional UNet architecture with the Diffusion Transformer (DiT), and employed flow-matching for improved training and sampling. However, they exhibit limited generation diversity. In this work, we leverage this limitation to perform consistent image edits via selective injection of attention features. The main challenge is that, unlike the UNet-based models, DiT lacks a coarse-to-fine synthesis structure, making it unclear in which layers to perform the injection. Therefore, we propose an automatic method to identify "vital layers" within DiT, crucial for image formation, and demonstrate how these layers facilitate a range of controlled stable edits, from non-rigid modifications to object addition, using the same mechanism. Next, to enable real-image editing, we introduce an improved image inversion method for flow models. Finally, we evaluate our approach through qualitative and quantitative comparisons, along with a user study, and demonstrate its effectiveness across multiple applications. The project page is available at https://omriavrahami.com/stable-flow

Related papers

• Taming Rectified Flow for Inversion and Editing [57.3742655030493]
  Rectified-flow-based diffusion transformers, such as FLUX and OpenSora, have demonstrated exceptional performance in the field of image and video generation. Despite their robust generative capabilities, these models often suffer from inaccurate inversion, which could limit their effectiveness in downstream tasks such as image and video editing. We propose RF-r, a novel training-free sampler that enhances inversion precision by reducing errors in the process of solving rectified flow ODEs.
  arXiv  Detail & Related papers  (2024-11-07T14:29:02Z)
• The Curious Case of End Token: A Zero-Shot Disentangled Image Editing using CLIP [4.710921988115686]
  We show that CLIP is capable of performing disentangled editing in a zero-shot manner. This insight may open opportunities for applying this method to various tasks, including image and video editing.
  arXiv  Detail & Related papers  (2024-06-01T14:46:57Z)
• Streamlining Image Editing with Layered Diffusion Brushes [8.738398948669609]
  Our system renders a single edit on a 512x512 image within 140 ms using a high-end consumer GPU. Our approach demonstrates efficacy across a range of tasks, including object attribute adjustments, error correction, and sequential prompt-based object placement and manipulation.
  arXiv  Detail & Related papers  (2024-05-01T04:30:03Z)
• Latent Space Editing in Transformer-Based Flow Matching [53.75073756305241]
  Flow Matching with a transformer backbone offers the potential for scalable and high-quality generative modeling. We introduce an editing space, $u$-space, that can be manipulated in a controllable, accumulative, and composable manner. Lastly, we put forth a straightforward yet powerful method for achieving fine-grained and nuanced editing using text prompts.
  arXiv  Detail & Related papers  (2023-12-17T21:49:59Z)
• BARET : Balanced Attention based Real image Editing driven by Target-text Inversion [36.59406959595952]
  We propose a novel editing technique that only requires an input image and target text for various editing types including non-rigid edits without fine-tuning diffusion model. Our method contains three novelties: (I) Targettext Inversion Schedule (TTIS) is designed to fine-tune the input target text embedding to achieve fast image reconstruction without image caption and acceleration of convergence; (II) Progressive Transition Scheme applies progressive linear approaches between target text embedding and its fine-tuned version to generate transition embedding for maintaining non-rigid editing capability; (III) Balanced Attention Module (BAM) balances the tradeoff between textual description and image semantics
  arXiv  Detail & Related papers  (2023-12-09T07:18:23Z)
• DragDiffusion: Harnessing Diffusion Models for Interactive Point-based Image Editing [94.24479528298252]
  DragGAN is an interactive point-based image editing framework that achieves impressive editing results with pixel-level precision. By harnessing large-scale pretrained diffusion models, we greatly enhance the applicability of interactive point-based editing on both real and diffusion-generated images. We present a challenging benchmark dataset called DragBench to evaluate the performance of interactive point-based image editing methods.
  arXiv  Detail & Related papers  (2023-06-26T06:04:09Z)
• Improving Diffusion-based Image Translation using Asymmetric Gradient Guidance [51.188396199083336]
  We present an approach that guides the reverse process of diffusion sampling by applying asymmetric gradient guidance. Our model's adaptability allows it to be implemented with both image-fusion and latent-dif models. Experiments show that our method outperforms various state-of-the-art models in image translation tasks.
  arXiv  Detail & Related papers  (2023-06-07T12:56:56Z)
• Cross-domain Compositing with Pretrained Diffusion Models [34.98199766006208]
  We employ a localized, iterative refinement scheme which infuses the injected objects with contextual information derived from the background scene. Our method produces higher quality and realistic results without requiring any annotations or training.
  arXiv  Detail & Related papers  (2023-02-20T18:54:04Z)
• End-to-End Visual Editing with a Generatively Pre-Trained Artist [78.5922562526874]
  We consider the targeted image editing problem: blending a region in a source image with a driver image that specifies the desired change. We propose a self-supervised approach that simulates edits by augmenting off-the-shelf images in a target domain. We show that different blending effects can be learned by an intuitive control of the augmentation process, with no other changes required to the model architecture.
  arXiv  Detail & Related papers  (2022-05-03T17:59:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.