郭新晨 - 博客园

2025年6月24日

SVAD: From Single Image to 3D Avatar via Synthetic Data Generation with Video Diffusion and Data Augmentation

摘要： https://yc4ny.github.io/SVAD/ 阅读全文

posted @ 2025-06-24 14:25 郭新晨阅读(14) 评论(0) 推荐(0)

2025年4月30日

Improving Subject-Driven Image Synthesis with Subject-Agnostic Guidance

摘要： https://openaccess.thecvf.com/content/CVPR2024/papers/Chan_Improving_Subject-Driven_Image_Synthesis_with_Subject-Agnostic_Guidance_CVPR_2024_paper.pdf 阅读全文

posted @ 2025-04-30 19:52 郭新晨阅读(43) 评论(0) 推荐(0)

Flux Already Knows – Activating Subject-Driven Image Generation without Training

摘要： https://arxiv.org/pdf/2504.11478 阅读全文

posted @ 2025-04-30 19:51 郭新晨阅读(89) 评论(0) 推荐(0)

PASTE, INPAINT AND HARMONIZE VIA DENOISING: SUBJECT-DRIVEN IMAGE EDITING WITH PRE-TRAINED DIFFUSION MODEL

摘要： https://arxiv.org/pdf/2306.07596 阅读全文

posted @ 2025-04-30 19:50 郭新晨阅读(25) 评论(0) 推荐(0)

DreamEdit: Subject-driven Image Editing

摘要： https://dreameditbenchteam.github.io/ 阅读全文

posted @ 2025-04-30 19:48 郭新晨阅读(51) 评论(0) 推荐(0)

Official implementations for paper: Customizable Image Synthesis with Multiple Subjects

摘要： https://github.com/ali-vilab/Cones-V2 阅读全文

posted @ 2025-04-30 19:47 郭新晨阅读(21) 评论(0) 推荐(0)

Cones 2: Customizable Image Synthesis with Multiple Subjects

摘要： https://cones-page.github.io/ 阅读全文

posted @ 2025-04-30 19:46 郭新晨阅读(24) 评论(0) 推荐(0)

UNO: A Universal Customization Method for Both Single and Multi-Subject Conditioning

摘要： https://github.com/bytedance/uno 阅读全文

posted @ 2025-04-30 19:45 郭新晨阅读(23) 评论(0) 推荐(0)

DynASyn: Multi-Subject Personalization Enabling Dynamic Action Synthesis

摘要： https://arxiv.org/abs/2503.17728 阅读全文

posted @ 2025-04-30 19:45 郭新晨阅读(18) 评论(0) 推荐(0)

探索多主体自定义图像合成新境界：Cones 2 深度解析与推荐

摘要： https://blog.csdn.net/gitblog_00214/article/details/141315562 阅读全文

posted @ 2025-04-30 19:44 郭新晨阅读(24) 评论(0) 推荐(0)

公告