ConsiStyle: Style Diversity in Training-Free Consistent T2I Generation

• URL: http://arxiv.org/abs/2505.20626v1
• Date: Tue, 27 May 2025 02:06:08 GMT
• Title: ConsiStyle: Style Diversity in Training-Free Consistent T2I Generation
• Authors: Yohai Mazuz, Janna Bruner, Lior Wolf,
• Abstract summary: We introduce a training-free method that achieves both style alignment and subject consistency.<n>Our method effectively decouples style from subject appearance and enables faithful generation of text-aligned images.
• Score: 49.1574468325115
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In text-to-image models, consistent character generation is the task of achieving text alignment while maintaining the subject's appearance across different prompts. However, since style and appearance are often entangled, the existing methods struggle to preserve consistent subject characteristics while adhering to varying style prompts. Current approaches for consistent text-to-image generation typically rely on large-scale fine-tuning on curated image sets or per-subject optimization, which either fail to generalize across prompts or do not align well with textual descriptions. Meanwhile, training-free methods often fail to maintain subject consistency across different styles. In this work, we introduce a training-free method that achieves both style alignment and subject consistency. The attention matrices are manipulated such that Queries and Keys are obtained from the anchor image(s) that are used to define the subject, while the Values are imported from a parallel copy that is not subject-anchored. Additionally, cross-image components are added to the self-attention mechanism by expanding the Key and Value matrices. To do without shifting from the target style, we align the statistics of the Value matrices. As is demonstrated in a comprehensive battery of qualitative and quantitative experiments, our method effectively decouples style from subject appearance and enables faithful generation of text-aligned images with consistent characters across diverse styles.

Related papers

• IP-Prompter: Training-Free Theme-Specific Image Generation via Dynamic Visual Prompting [71.29100512700064]
  IP-Prompter is a novel training-free TSI generation method.<n>It integrates reference images into generative models, allowing users to seamlessly specify the target theme.<n>Our approach enables diverse applications, including consistent story generation, character design, realistic character generation, and style-guided image generation.
  arXiv  Detail & Related papers  (2025-01-26T19:01:19Z)
• Beyond Color and Lines: Zero-Shot Style-Specific Image Variations with Coordinated Semantics [3.9717825324709413]
  Style has been primarily considered in terms of artistic elements such as colors, brushstrokes, and lighting. In this study, we propose a zero-shot scheme for image variation with coordinated semantics.
  arXiv  Detail & Related papers  (2024-10-24T08:34:57Z)
• StyleForge: Enhancing Text-to-Image Synthesis for Any Artistic Styles with Dual Binding [7.291687946822539]
  We introduce Single-StyleForge, a novel approach for personalized text-to-image synthesis across diverse artistic styles. We also present Multi-StyleForge, which enhances image quality and text alignment by binding multiple tokens to partial style attributes.
  arXiv  Detail & Related papers  (2024-04-08T07:43:23Z)
• InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation [5.364489068722223]
  The concept of style is inherently underdetermined, encompassing a multitude of elements such as color, material, atmosphere, design, and structure. Inversion-based methods are prone to style degradation, often resulting in the loss of fine-grained details. adapter-based approaches frequently require meticulous weight tuning for each reference image to achieve a balance between style intensity and text controllability.
  arXiv  Detail & Related papers  (2024-04-03T13:34:09Z)
• Training-Free Consistent Text-to-Image Generation [80.4814768762066]
  Text-to-image models can portray the same subject across diverse prompts. Existing approaches fine-tune the model to teach it new words that describe specific user-provided subjects. We present ConsiStory, a training-free approach that enables consistent subject generation by sharing the internal activations of the pretrained model.
  arXiv  Detail & Related papers  (2024-02-05T18:42:34Z)
• Pick-and-Draw: Training-free Semantic Guidance for Text-to-Image Personalization [56.12990759116612]
  Pick-and-Draw is a training-free semantic guidance approach to boost identity consistency and generative diversity for personalization methods. The proposed approach can be applied to any personalized diffusion models and requires as few as a single reference image.
  arXiv  Detail & Related papers  (2024-01-30T05:56:12Z)
• Style Aligned Image Generation via Shared Attention [61.121465570763085]
  We introduce StyleAligned, a technique designed to establish style alignment among a series of generated images. By employing minimal attention sharing' during the diffusion process, our method maintains style consistency across images within T2I models. Our method's evaluation across diverse styles and text prompts demonstrates high-quality and fidelity.
  arXiv  Detail & Related papers  (2023-12-04T18:55:35Z)
• MSSRNet: Manipulating Sequential Style Representation for Unsupervised Text Style Transfer [82.37710853235535]
  Unsupervised text style transfer task aims to rewrite a text into target style while preserving its main content. Traditional methods rely on the use of a fixed-sized vector to regulate text style, which is difficult to accurately convey the style strength for each individual token. Our proposed method addresses this issue by assigning individual style vector to each token in a text, allowing for fine-grained control and manipulation of the style strength.
  arXiv  Detail & Related papers  (2023-06-12T13:12:29Z)
• Domain Enhanced Arbitrary Image Style Transfer via Contrastive Learning [84.8813842101747]
  Contrastive Arbitrary Style Transfer (CAST) is a new style representation learning and style transfer method via contrastive learning. Our framework consists of three key components, i.e., a multi-layer style projector for style code encoding, a domain enhancement module for effective learning of style distribution, and a generative network for image style transfer.
  arXiv  Detail & Related papers  (2022-05-19T13:11:24Z)
• STALP: Style Transfer with Auxiliary Limited Pairing [36.23393954839379]
  We present an approach to example-based stylization of images that uses a single pair of a source image and its stylized counterpart. We demonstrate how to train an image translation network that can perform real-time semantically meaningful style transfer to a set of target images.
  arXiv  Detail & Related papers  (2021-10-20T11:38:41Z)

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.