图像生成 - arXivDaily 专题

2512.19048 2025-12-23 cs.CV 70%

WaTeRFlow: Watermark Temporal Robustness via Flow Consistency

WaTeRFlow：通过流一致性实现水印时间鲁棒性

Utae Jeong, Sumin In, Hyunju Ryu, Jaewan Choi, Feng Yang, Jongheon Jeong, Seungryong Kim, Sangpil Kim

机构 * Korea University（韩国大学）； Google DeepMind（谷歌DeepMind）； KAIST AI（韩国科学技术院人工智能）

专题命中个性化与一致性：diffusion(abstract)；image editing(abstract)；分类 cs.CV

AI总结 WaTeRFlow通过流一致性提升I2V场景下的水印鲁棒性，结合FUSE引擎、时间一致性损失和语义保持损失，实现跨模态水印恢复。

2512.09417 2025-12-11 cs.CV 70%

DirectSwap: Mask-Free Cross-Identity Training and Benchmarking for Expression-Consistent Video Head Swapping

DirectSwap: 无掩码跨身份训练与基准测试用于表情一致的视频头部交换

Yanan Wang, Shengcai Liao, Panwen Hu, Xin Li, Fan Yang, Xiaodan Liang

机构 * MBZUAI（马克斯·普朗克智能研究院）； UAEU（阿拉伯联合酋长国大学）； AIQ（人工智能量子研究所）； SYSU（南方大学）

专题命中个性化与一致性：diffusion(abstract)；inpainting(abstract)；分类 cs.CV

AI总结 DirectSwap通过无掩码直接视频头部交换框架，结合运动和表情感知重建损失，实现了跨身份视频头部交换的高质量和一致性。

2511.17344 2025-12-01 cs.CV 70%

Loomis Painter: Reconstructing the Painting Process

Loomis Painter: 复原绘画过程

Markus Pobitzer, Chang Liu, Chenyi Zhuang, Teng Long, Bin Ren, Nicu Sebe

机构 * University of Trento（特伦托大学）； University of Pisa（比萨大学）

专题命中个性化与一致性：diffusion(abstract)；image synthesis(abstract)；分类 cs.CV

AI总结 Loomis Painter通过统一框架和语义驱动的风格控制机制，实现多媒体绘画过程的复原，提升跨媒体一致性与生成流畅性。

2511.06721 2025-11-11 cs.CV 70%

AvatarTex: High-Fidelity Facial Texture Reconstruction from Single-Image Stylized Avatars

Yuda Qiu, Zitong Xiao, Yiwei Zuo, Zisheng Ye, Weikai Chen, Xiaoguang Han

机构 * SSE, CUHKSZ（CUHKSZ 信息学院）； FNii, CUHKSZ（CUHKSZ 信息学院）

专题命中个性化与一致性：diffusion(abstract)；inpainting(abstract)；分类 cs.CV

Comments 3DV 2026 Accepted

2511.01355 2025-11-04 cs.CV 70%

Expanding the Content-Style Frontier: a Balanced Subspace Blending Approach for Content-Style LoRA Fusion

Linhao Huang

机构 * Beijing University of Technology（北京理工大学）

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

2510.22994 2025-10-28 cs.CV 70%

SceneDecorator: Towards Scene-Oriented Story Generation with Scene Planning and Scene Consistency

Quanjian Song, Donghao Zhou, Jingyu Lin, Fei Shen, Jiaze Wang, Xiaowei Hu, Cunjian Chen, Pheng-Ann Heng

机构 * Monash University（莫纳什大学）； The Chinese University of Hong Kong（香港中文大学）； National University of Singapore（新加坡国立大学）； South China University of Technology（华南理工大学）

专题命中个性化与一致性：image generation(abstract)；text-to-image(abstract)；分类 cs.CV

Comments Accepted by NeurIPS 2025; Project Page: https://lulupig12138.github.io/SceneDecorator

2510.20887 2025-10-27 cs.CV cs.AI 70%

Preventing Shortcuts in Adapter Training via Providing the Shortcuts

Anujraaj Argo Goyal, Guocheng Gordon Qian, Huseyin Coskun, Aarush Gupta, Himmy Tam, Daniil Ostashev, Ju Hu, Dhritiman Sagar, Sergey Tulyakov, Kfir Aberman, Kuan-Chieh Jackson Wang

机构 * Snap Inc.

专题命中个性化与一致性：text-to-image(abstract)；image synthesis(abstract)；分类 cs.CV

Comments Accepted to NeurIPS 2025, webpage: https://snap-research.github.io/shortcut-rerouting/

2510.16791 2025-10-21 cs.CV 70%

Personalized Image Filter: Mastering Your Photographic Style

Chengxuan Zhu, Shuchen Weng, Jiacong Fang, Peixuan Zhang, Si Li, Chao Xu, Boxin Shi

机构 * State Key Laboratory of General Artificial Intelligence, School of Intelligence Science and Technology, Peking University（国家通用人工智能重点实验室，智能科学与技术学院，北京大学）； Beijing Academy of Artificial Intelligence（北京人工智能研究院）； State Key Laboratory of Multimedia Information Processing and National Engineering Research Center of Visual Technology, School of Computer Science, Peking University（多媒体信息处理国家重点实验室和视觉技术国家工程研究中心，计算机科学学院，北京大学）； School of Artificial Intelligence, Beijing University of Posts and Telecommunications（人工智能学院，北京邮电大学）

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

2509.20775 2025-09-26 cs.CV cs.AI 70%

CusEnhancer: A Zero-Shot Scene and Controllability Enhancement Method for Photo Customization via ResInversion

Maoye Ren, Praneetha Vaddamanu, Jianjin Xu, Fernando De la Torre Frade

机构 * College of Computer Science and Technology, East China University of Science and Technology（计算机科学与技术学院，东华大学）； Carnegie Mellon University（卡内基梅隆大学）； Microsoft（微软公司）

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

2509.17120 2025-09-23 cs.CV 70%

Stencil: Subject-Driven Generation with Context Guidance

Gordon Chen, Ziqi Huang, Cheston Tan, Ziwei Liu

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

Comments Accepted as Spotlight at ICIP 2025

Journal ref Proc. IEEE Int. Conf. Image Process. (ICIP), Anchorage, AK, USA, Sept. 14-17, 2025, pp. 719-724

2509.17088 2025-09-23 cs.CV 70%

AlignedGen: Aligning Style Across Generated Images

Jiexuan Zhang, Yiheng Du, Qian Wang, Weiqi Li, Yu Gu, Jian Zhang

机构 * School of Electronic and Computer Engineering, Peking University（电子与计算机工程学院，北京大学）

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

2509.10259 2025-09-15 cs.CV 70%

Mask Consistency Regularization in Object Removal

Hua Yuan, Jin Yuan, Yicheng Jiang, Yao Zhang, Xin Geng, Yong Rui

机构 * School of Computer Science and Engineering, Southeast University（东南大学计算机科学与工程学院）； AI Lab, Lenovo Research（联想研究院人工智能实验室）

专题命中个性化与一致性：diffusion(abstract)；inpainting(abstract)；分类 cs.CV

2504.12574 2025-09-10 cs.CV 70%

ForgetMe: Evaluating Selective Forgetting in Generative Models

Zhenyu Yu, Mohd Yamani Inda Idris, Pei Wang

机构 * Universiti Malaya（马来亚大学）； Kunming University of Science and Technology（昆明理工大学）

专题命中个性化与一致性：image generation(abstract)；diffusion(abstract)；分类 cs.CV

2503.12615 2025-09-04 cs.CV cs.LG 70%

LATINO-PRO: LAtent consisTency INverse sOlver with PRompt Optimization

Alessio Spagnoletti, Jean Prost, Andrés Almansa, Nicolas Papadakis, Marcelo Pereyra

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

Comments 27 pages, 24 figures, International Conference on Computer Vision, ICCV 2025

2503.06397 2025-09-03 cs.CV 70%

Removing Averaging: Personalized Lip-Sync Driven Characters Based on Identity Adapter

Yanyu Zhu, Lichen Bai, Jintao Xu, Hai-tao Zheng

机构 * Shenzhen International Graduate School, Tsinghua Universiy（清华大学深圳国际研究生院）； Tencent（腾讯）； Seoul National University（首尔国立大学）

专题命中个性化与一致性：diffusion(abstract)；inpainting(abstract)；分类 cs.CV

2508.11284 2025-08-19 cs.CV 70%

TimeMachine: Fine-Grained Facial Age Editing with Identity Preservation

Yilin Mi, Qixin Yan, Zheng-Peng Duan, Chunle Guo, Hubery Yin, Hao Liu, Chen Li, Chongyi Li

专题命中个性化与一致性：diffusion(abstract)；image editing(abstract)；分类 cs.CV

2508.11624 2025-08-18 cs.CV 70%

LoRAtorio: An intrinsic approach to LoRA Skill Composition

Niki Foteinopoulou, Ignas Budvytis, Stephan Liwicki

机构 * Cambridge Research Laboratory, Toshiba Europe（东芝欧洲 Cambridge 研究实验室）

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

Comments 32 pages, 17 figures

2508.03241 2025-08-07 cs.CV 70%

FFHQ-Makeup: Paired Synthetic Makeup Dataset with Facial Consistency Across Multiple Styles

Xingchao Yang, Shiori Ueda, Yuantian Huang, Tomoya Akiyama, Takafumi Taketomi

机构 * CyberAgent ； Keio University（Keio大学）

专题命中个性化与一致性：image generation(abstract)；text-to-image(abstract)；分类 cs.CV

Comments Project: https://yangxingchao.github.io/FFHQ-Makeup-page, Datasets: https://huggingface.co/datasets/cyberagent/FFHQ-Makeup

2507.11986 2025-08-06 cs.CV 70%

Style Composition within Distinct LoRA modules for Traditional Art

Jaehyun Lee, Wonhark Park, Wonsik Shin, Hyunho Lee, Hyoung Min Na, Nojun Kwak

机构 * Seoul National University（首尔国立大学）； Kyunghee University（庆熙大学）

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

Comments Accepted to ICCV 2025 Workshop(WCCA)

2508.01684 2025-08-05 cs.CV 70%

DisCo3D: Distilling Multi-View Consistency for 3D Scene Editing

Yufeng Chi, Huimin Ma, Kafeng Wang, Jianmin Li

机构 * Institute for Artificial Intelligence, Beijing National Research Center for Information Science and Technology, Department of Computer Science and Technology, Tsinghua University（人工智能研究院、北京信息科学与技术国家研究中心、计算机科学与技术系、清华大学）； College of Computer, National University of Defense Technology（计算机学院、国防科技大学）

专题命中个性化与一致性：image generation(abstract)；diffusion(abstract)；分类 cs.CV

Comments 17 pages, 7 figures

2504.02612 2025-07-31 cs.CV 70%

Fine-Tuning Visual Autoregressive Models for Subject-Driven Generation

Jiwoo Chung, Sangeek Hyun, Hyunjun Kim, Eunseo Koh, MinKyu Lee, Jae-Pil Heo

机构 * Sungkyunkwan University（成均馆大学）

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

Comments Accepted to ICCV 2025. Project page: https://jiwoogit.github.io/ARBooth/

2503.16418 2025-07-30 cs.CV cs.LG 70%

InfiniteYou: Flexible Photo Recrafting While Preserving Your Identity

Liming Jiang, Qing Yan, Yumin Jia, Zichuan Liu, Hao Kang, Xin Lu

机构 * ByteDance Intelligent Creation Project（字节跳动智能创作项目）

专题命中个性化与一致性：image generation(abstract)；diffusion(abstract)；分类 cs.CV

Comments ICCV 2025 (Highlight). Project page: https://bytedance.github.io/InfiniteYou/ Code and model: https://github.com/bytedance/InfiniteYou

2507.11533 2025-07-16 cs.CV 70%

CharaConsist: Fine-Grained Consistent Character Generation

Mengyu Wang, Henghui Ding, Jianing Peng, Yao Zhao, Yunpeng Chen, Yunchao Wei

机构 * Institute of Information Science, Beijing Jiaotong University（信息科学学院，北京交通大学）； Institute of Big Data, Fudan University（大数据学院，复旦大学）； Visual Intelligence + X International Joint Laboratory of the Ministry of Education（教育部视觉智能+X国际联合实验室）； Alkaid Pte. Ltd., Singapore（新加坡Alkaid公司）

专题命中个性化与一致性：image generation(abstract)；text-to-image(abstract)；分类 cs.CV

Comments ICCV 2025 accepted paper, project page: https://murray-wang.github.io/CharaConsist/

2412.00136 2025-07-14 cs.CV cs.AI cs.LG 70%

FonTS: Text Rendering with Typography and Style Controls

Wenda Shi, Yiren Song, Dengming Zhang, Jiaming Liu, Xingxing Zou

机构 * The Hong Kong Polytechnic University（香港理工大学）； National University of Singapore（新加坡国立大学）； Zhejiang University（浙江大学）； Tiamat AI

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

Comments Accepted to ICCV 2025

2507.00789 2025-07-02 cs.CV 70%

OptiPrune: Boosting Prompt-Image Consistency with Attention-Guided Noise and Dynamic Token Selection

Ziji Lu

机构 * School of Computational Science \& Education Software, Guangzhou University, 510006, Guangzhou, PR China

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

2503.22352 2025-07-01 cs.CV 70%

Meta-LoRA: Meta-Learning LoRA Components for Domain-Aware ID Personalization

Barış Batuhan Topal, Umut Özyurt, Zafer Doğan Budak, Ramazan Gokberk Cinbis

机构 * Pixery Labs İstanbul, Türkiye（Pixery实验室伊斯坦布尔，土耳其）； METU Dept. of Computer Engineering（METU计算机工程系）； METU Dept. of Electrical and Elec. Engineering（METU电气与电子工程系）

专题命中个性化与一致性：text-to-image(abstract)；diffusion(abstract)；分类 cs.CV

2506.07750 2025-06-10 cs.CV 70%

Difference Inversion: Interpolate and Isolate the Difference with Token Consistency for Image Analogy Generation

Hyunsoo Kim, Donghyun Kim, Suhyun Kim

机构 * Korea University（韩国大学）； Korea Institute of Science and Technology（韩国科学技术院）； Kyung Hee University（庆熙大学）

专题命中个性化与一致性：diffusion(abstract)；inpainting(abstract)；分类 cs.CV

Comments Published at CVPR 2025

2502.18197 2025-06-05 cs.LG cs.CV 70%

VCT: Training Consistency Models with Variational Noise Coupling

Gianluigi Silvestri, Luca Ambrogioni, Chieh-Hsin Lai, Yuhta Takida, Yuki Mitsufuji

专题命中个性化与一致性：image generation(abstract)；diffusion(abstract)；分类 cs.CV

Comments 23 pages, 11 figures

2504.01014 2025-06-02 cs.CV 70%

AnimeGamer: Infinite Anime Life Simulation with Next Game State Prediction

Junhao Cheng, Yuying Ge, Yixiao Ge, Jing Liao, Ying Shan

机构 * ARC Lab, Tencent PCG（腾讯PCG ARC实验室）； City University of Hong Kong（香港城市大学）

专题命中个性化与一致性：image generation(abstract)；diffusion(abstract)；分类 cs.CV

Comments Project released at: https://howe125.github.io/AnimeGamer.github.io/

2505.14843 2025-05-22 cs.CV cs.AI 70%

Leveraging Generative AI Models to Explore Human Identity

Yunha Yeo, Daeho Um

机构 * Graduate School of Culture Technology, KAIST（韩国科学技术院文化科技研究生院）； Samsung Advanced Institute of Technology (SAIT)（三星先进技术研究所）

专题命中个性化与一致性：image generation(abstract)；diffusion(abstract)；分类 cs.CV

Comments Accepted to ISEA 2025

1. 个性化与一致性 1266 篇

WaTeRFlow: Watermark Temporal Robustness via Flow Consistency

DirectSwap: Mask-Free Cross-Identity Training and Benchmarking for Expression-Consistent Video Head Swapping

Loomis Painter: Reconstructing the Painting Process

AvatarTex: High-Fidelity Facial Texture Reconstruction from Single-Image Stylized Avatars

Expanding the Content-Style Frontier: a Balanced Subspace Blending Approach for Content-Style LoRA Fusion

SceneDecorator: Towards Scene-Oriented Story Generation with Scene Planning and Scene Consistency

Preventing Shortcuts in Adapter Training via Providing the Shortcuts

Personalized Image Filter: Mastering Your Photographic Style

CusEnhancer: A Zero-Shot Scene and Controllability Enhancement Method for Photo Customization via ResInversion

Stencil: Subject-Driven Generation with Context Guidance

AlignedGen: Aligning Style Across Generated Images

Mask Consistency Regularization in Object Removal

ForgetMe: Evaluating Selective Forgetting in Generative Models

LATINO-PRO: LAtent consisTency INverse sOlver with PRompt Optimization

Removing Averaging: Personalized Lip-Sync Driven Characters Based on Identity Adapter

TimeMachine: Fine-Grained Facial Age Editing with Identity Preservation

LoRAtorio: An intrinsic approach to LoRA Skill Composition

FFHQ-Makeup: Paired Synthetic Makeup Dataset with Facial Consistency Across Multiple Styles

Style Composition within Distinct LoRA modules for Traditional Art

DisCo3D: Distilling Multi-View Consistency for 3D Scene Editing

Fine-Tuning Visual Autoregressive Models for Subject-Driven Generation

InfiniteYou: Flexible Photo Recrafting While Preserving Your Identity

CharaConsist: Fine-Grained Consistent Character Generation

FonTS: Text Rendering with Typography and Style Controls

OptiPrune: Boosting Prompt-Image Consistency with Attention-Guided Noise and Dynamic Token Selection

Meta-LoRA: Meta-Learning LoRA Components for Domain-Aware ID Personalization

Difference Inversion: Interpolate and Isolate the Difference with Token Consistency for Image Analogy Generation

VCT: Training Consistency Models with Variational Noise Coupling

AnimeGamer: Infinite Anime Life Simulation with Next Game State Prediction

Leveraging Generative AI Models to Explore Human Identity