paper-with-me

Papers

StyleBlend: Enhancing Style-Specific Content Creation in Text-to-Image Diffusion Models

2025-02-13 · Zichong Chen, Shijin Wang, Yang Zhou

Synthesizing visually impressive images that seamlessly align both text prompts and specific artistic styles remains a significant challenge in Text-to-Image (T2I) diffusion models. This paper introduces StyleBlend, a method designed to learn and apply style representations from a limited set of reference images, enabling content synthesis of both text-aligned and stylistically coherent. Our approach uniquely decomposes style into two components, composition and texture, each learned through different strategies. We then leverage two synthesis branches, each focusing on a corresponding style component, to facilitate effective style blending through shared features without affecting content generation. StyleBlend addresses the common issues of text misalignment and weak style representation that previous methods have struggled with. Extensive qualitative and quantitative comparisons demonstrate the superiority of our approach.

📄 PDF Abstract BibTeX arXiv:2502.09064

Code (1)

zichongc/StyleBlend 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…
ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…
SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Style-Friendly SNR Sampler for Style-Driven Generation

2024-11-22 · Jooyoung Choi, Chaehun Shin, Yeongtak Oh, Heeseung Kim 외

Recent large-scale diffusion models generate high-quality images but struggle to learn new, personalized artistic styles, which limits the creation of unique style templates. Fine-tuning with reference images is the most…

GPU-Accelerated Mobile Multi-view Style Transfer

2020-03-02 · Puneet Kohli, Saravana Gunaseelan, Jason Orozco, Yiwen Hua 외

An estimated 60% of smartphones sold in 2018 were equipped with multiple rear cameras, enabling a wide variety of 3D-enabled applications such as 3D Photos. The success of 3D Photo platforms (Facebook 3D Photo, Holopix, …

GPUImage InpaintingImage ManipulationLearning Representation Of Multi-View Data+3

PARASOL: Parametric Style Control for Diffusion Image Synthesis

2023-03-11 · Gemma Canet Tarrés, Dan Ruta, Tu Bui, John Collomosse

We propose PARASOL, a multi-modal synthesis model that enables disentangled, parametric control of the visual style of the image by jointly conditioning synthesis on both content and a fine-grained visual style embedding…

Image Generation

Integrating Randomness in Large Language Models: A Linear Congruential Generator Approach for Generating Clinically Relevant Content

2024-07-04 · Andrew Bouras

Generating diverse, high-quality outputs from language models is crucial for applications in education and content creation. Achieving true randomness and avoiding repetition remains a significant challenge. This study u…

Fact SelectionLanguage ModelingLanguage Modelling

ControlStyle: Text-Driven Stylized Image Generation Using Diffusion Priors

2023-11-09 · Jingwen Chen, Yingwei Pan, Ting Yao, Tao Mei

Recently, the multimedia community has witnessed the rise of diffusion models trained on large-scale multi-modal data for visual content creation, particularly in the field of text-to-image generation. In this paper, we …

Image GenerationStyle TransferText to Image GenerationText-to-Image Generation