3D 视觉 - arXivDaily 专题

2608.01825 2026-08-04 cs.CV cs.AI cs.GR 新提交 62%

PartMat: Material-Aware 3D Part Decomposition with a Single Global Latent

PartMat：基于单个全局隐变量的材质感知三维部件分解

Guangming Fu, Jin Song, Yiyun Fei, Guoqiu Li, Ruigao Yang, Jianan Jiang

机构 * Alibaba Group（阿里巴巴集团）

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

AI总结该研究提出PartMat，一种用单个全局隐变量的材质感知三维部件分解流水线，可按材质边界分解物体，推理高效且分解准确率优于基线，几何质量相当。

2605.29655 2026-07-08 cs.CV cs.GR 版本更新 62%

SuperVoxelGPT: Adaptive and Ordered 3D Tokenization for Autoregressive Shape Generation

SuperVoxelGPT: 自适应有序3D令牌化用于自回归形状生成

Yuan Li, Congyi Zhang, Xifeng Gao, Xiaohu Guo

机构 * University of Texas at Dallas（德克萨斯大学达拉斯分校）； Tencent America（腾讯美国）

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

AI总结提出SuperVoxelGPT框架，通过自适应且有序的超体素令牌化解决自回归3D生成中序列长度与空间顺序的矛盾，实现高质量、高效率的形状生成。

2606.28215 2026-06-29 cs.CV cs.AI cs.GR 新提交 62%

HAT-4D: Lifting Monocular Video for 4D Multi-Object Interactions via Human-Agent Collaboration

HAT-4D: 通过人机协作从单目视频提升4D多物体交互

Jiaxin Li, Yuxiang Wu, Zhenkai Zhang, Xinrui Shi, Haoyuan Wang, Yichen Zhao, Su Linxiang, Chenyang Yu, Mingyu Zhang, Yifan Ding, Boran Wen, Li Zhang, Ruiyang Liu, Yong-Lu Li

机构 * Shanghai Jiao Tong University（上海交通大学）； Shanghai Innovation Institute（上海创新研究院）； University of Science and Technology of China（中国科学技术大学）； Math Magic

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

AI总结提出HAT-4D框架，结合视觉大模型与多级人工反馈，从单目视频重建多物体的3D几何、时序动态和物理交互，解决遮挡和深度歧义，无需多相机系统。

Comments Accepted to ECCV 2026. 15 pages of main text and 39 pages of appendices. Project page: https://lijiaxin0111.github.io/HAT4D/

2606.23514 2026-06-23 cs.CV cs.GR 新提交 62%

Arbor: Explicit Geometric Conditioning for Controllable 3D Asset Generation

Arbor: 用于可控3D资产生成的显式几何条件

Jan-Niklas Dihlmann, Andreas Engelhardt, Simon Donne, Hendrik P. A. Lensch, Mark Boss

机构 * University of Tübingen（图宾根大学）； Stability AI

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

AI总结提出Arbor，通过约束网格作为原生3D控制接口，在冻结去噪器中学习路由附件，实现文本条件3D生成中空间约束的显式控制，提升约束服从性。

Comments Project Page: https://arbor.jdihlmann.com/

2605.26137 2026-05-27 cs.GR cs.AI cs.CV 62%

AssetGen: Deployable 3D Asset Generation at Interactive Speed

AssetGen: 可部署的交互速度3D资产生成

Dilin Wang, Xiaoyu Xiang, Kihyuk Sohn, Tom Monnier, Yu-Ying Yeh, Thu Nguyen-Phuoc, Jiawen Zhang, Yuchen Fan, Antoine Toisoul, Hyunyoung Jung, Prithviraj Dhar, Michael Bunnell, Nikolaos Sarafianos, Chuhang Zou, Roman Shapovalov, Andrea Vedaldi, Rakesh Ranjan

机构 * Reality Labs, Meta（Meta现实实验室）

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

AI总结提出AssetGen系统，通过粗到细的VecSet框架、多视图纹理生成及端到端加速，在30秒内生成带烘焙法线、颜色纹理和可控多边形预算的高质量网格，支持实时渲染和移动端部署。

2603.29089 2026-04-01 cs.CV cs.AI cs.GR 62%

WorldFlow3D: Flowing Through 3D Distributions for Unbounded World Generation

WorldFlow3D: 通过3D分布流动实现无界世界生成

Amogh Joshi, Julian Ost, Felix Heide

机构 * Princeton University（普林斯顿大学）； Torc Robotics

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

AI总结 WorldFlow3D通过流动3D分布生成无界3D世界，具备快速收敛、高精度结构与纹理生成能力，并在真实和合成场景中验证了跨领域泛化性。

2504.11734 2025-12-23 cs.GR cs.CV 62%

Recent Advances in 3D Object and Scene Generation: A Survey

近年来3D对象和场景生成的最新进展：综述

Xiang Tang, Ruotong Li, Xiaopeng Fan

机构 * Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳））； Pengcheng Laboratory（鹏城实验室）； Harbin Institute of Technology（哈尔滨工业大学）

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

AI总结本文综述了3D对象和场景生成的最新进展，分析了主流生成模型和技术路径，指出了当前挑战与未来研究方向。

Comments 35 pages, 7 figures, 6 tables, Project page: https://github.com/xdlbw/Awesome-3D-Object-and-Scene-Generation

2511.13648 2025-11-18 cs.CV cs.RO 62%

PhysX-Anything: Simulation-Ready Physical 3D Assets from Single Image

Ziang Cao, Fangzhou Hong, Zhaoxi Chen, Liang Pan, Ziwei Liu

机构 * S-Lab, Nanyang Technological University（南洋理工大学S实验室）； Shanghai AI Lab（上海人工智能实验室）

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.RO

Comments Project page: https://physx-anything.github.io/

2510.25765 2025-11-05 cs.CV cs.GR 62%

FreeArt3D: Training-Free Articulated Object Generation using 3D Diffusion

Chuhao Chen, Isabella Liu, Xinyue Wei, Hao Su, Minghua Liu

机构 * University of California San Diego（加州大学圣地亚哥分校）； Hillbot Inc.（Hillbot公司）

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments Project Page: https://czzzzh.github.io/FreeArt3D Code: https://github.com/CzzzzH/FreeArt3D

2507.11465 2025-07-16 cs.GR cs.CV 62%

Elevating 3D Models: High-Quality Texture and Geometry Refinement from a Low-Quality Model

Nuri Ryu, Jiyun Won, Jooeun Son, Minsu Gong, Joo-Haeng Lee, Sunghyun Cho

机构 * POSTECH ； Pebblous

专题命中 3D生成：3D vision(abstract)；分类 cs.CV、cs.GR

Comments Accepted to SIGGRAPH 2025. For the project page, see https://cg.postech.ac.kr/research/Elevate3D/

2411.17945 2025-03-27 cs.CV cs.AI cs.GR cs.LG 62%

MARVEL-40M+: Multi-Level Visual Elaboration for High-Fidelity Text-to-3D Content Creation

Sankalp Sinha, Mohammad Sadil Khan, Muhammad Usama, Shino Sam, Didier Stricker, Sk Aziz Ali, Muhammad Zeshan Afzal

专题命中 3D生成：3D reconstruction(abstract)；分类 cs.CV、cs.GR

2503.01370 2025-03-24 cs.GR cs.CV cs.MM 62%

Kiss3DGen: Repurposing Image Diffusion Models for 3D Asset Generation

Jiantao Lin, Xin Yang, Meixi Chen, Yingjie Xu, Dongyu Yan, Leyi Wu, Xinli Xu, Lie XU, Shunsi Zhang, Ying-Cong Chen

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments The first three authors contributed equally to this work

2502.15869 2025-02-25 cs.GR cs.AI cs.CV cs.HC 62%

Generative AI Framework for 3D Object Generation in Augmented Reality

Majid Behravan

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

2502.10606 2025-02-18 cs.CV cs.RO 62%

HIPPo: Harnessing Image-to-3D Priors for Model-free Zero-shot 6D Pose Estimation

Yibo Liu, Zhaodong Jiang, Binbin Xu, Guile Wu, Yuan Ren, Tongtong Cao, Bingbing Liu, Rui Heng Yang, Amir Rasouli, Jinjun Shan

专题命中 3D生成：3D reconstruction(abstract)；分类 cs.CV、cs.RO

2501.16177 2025-01-28 cs.CV cs.AI cs.GR 62%

BAG: Body-Aligned 3D Wearable Asset Generation

Zhongjin Luo, Yang Li, Mingrui Zhang, Senbo Wang, Han Yan, Xibin Song, Taizhang Shang, Wei Mao, Hongdong Li, Xiaoguang Han, Pan Ji

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments video: https://youtu.be/XJtG82LjQKc

2412.15200 2024-12-20 cs.CV cs.AI cs.GR 62%

DI-PCG: Diffusion-based Efficient Inverse Procedural Content Generation for High-quality 3D Asset Creation

Wang Zhao, Yan-Pei Cao, Jiale Xu, Yuejiang Dong, Ying Shan

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments Project page: https://thuzhaowang.github.io/projects/DI-PCG/

2412.01106 2024-12-03 cs.CV cs.GR 62%

One Shot, One Talk: Whole-body Talking Avatar from a Single Image

Jun Xiang, Yudong Guo, Leipeng Hu, Boyang Guo, Yancheng Yuan, Juyong Zhang

专题命中 3D生成：3DGS(abstract)；分类 cs.CV、cs.GR

Comments Project Page: https://ustc3dv.github.io/OneShotOneTalk/

2411.04928 2024-11-08 cs.CV cs.AI cs.GR 62%

DimensionX: Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion

Wenqiang Sun, Shuo Chen, Fangfu Liu, Zilong Chen, Yueqi Duan, Jun Zhang, Yikai Wang

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments Project Page: https://chenshuo20.github.io/DimensionX/

2405.15891 2024-10-11 cs.CV cs.GR cs.LG 62%

Score Distillation via Reparametrized DDIM

Artem Lukoianov, Haitz Sáez de Ocáriz Borde, Kristjan Greenewald, Vitor Campagnolo Guizilini, Timur Bagautdinov, Vincent Sitzmann, Justin Solomon

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments NeurIPS 2024. 28 pages, 30 figures. Revision: additional comparisons and ablations studies

2403.03954 2024-09-30 cs.RO cs.CV cs.LG 62%

3D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations

Yanjie Ze, Gu Zhang, Kangning Zhang, Chenyuan Hu, Muhan Wang, Huazhe Xu

专题命中 3D生成：point cloud(abstract)；分类 cs.CV、cs.RO

Comments Published at Robotics: Science and Systems (RSS) 2024. Videos, code, and data: https://3d-diffusion-policy.github.io

2403.16803 2024-09-17 cs.RO cs.CV 62%

Exploiting Priors from 3D Diffusion Models for RGB-Based One-Shot View Planning

Sicong Pan, Liren Jin, Xuying Huang, Cyrill Stachniss, Marija Popović, Maren Bennewitz

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.RO

Comments Sicong Pan and Liren Jin have equal contribution. Publication to appear in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024

2405.09431 2024-05-16 cs.CV cs.GR 62%

A Survey On Text-to-3D Contents Generation In The Wild

Chenhan Jiang

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments 11 pages, 10 figures, 4 tables. arXiv admin note: text overlap with arXiv:2401.17807 by other authors

2405.08054 2024-05-15 cs.GR cs.CV 62%

Coin3D: Controllable and Interactive 3D Assets Generation with Proxy-Guided Conditioning

Wenqi Dong, Bangbang Yang, Lin Ma, Xiao Liu, Liyuan Cui, Hujun Bao, Yuewen Ma, Zhaopeng Cui

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments Project webpage: https://zju3dv.github.io/coin3d

2310.08529 2024-05-14 cs.CV cs.GR 62%

GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models

Taoran Yi, Jiemin Fang, Junjie Wang, Guanjun Wu, Lingxi Xie, Xiaopeng Zhang, Wenyu Liu, Qi Tian, Xinggang Wang

专题命中 3D生成：Gaussian Splatting(abstract)；分类 cs.CV、cs.GR

Comments CVPR 2024, Project page: https://taoranyi.com/gaussiandreamer/

2309.03453 2024-04-16 cs.CV cs.AI cs.GR 62%

SyncDreamer: Generating Multiview-consistent Images from a Single-view Image

Yuan Liu, Cheng Lin, Zijiao Zeng, Xiaoxiao Long, Lingjie Liu, Taku Komura, Wenping Wang

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments ICLR 2024 Spotlight. Project page: https://liuyuan-pal.github.io/SyncDreamer/ Code: https://github.com/liuyuan-pal/SyncDreamer

2403.12032 2024-03-20 cs.CV cs.GR 62%

Generic 3D Diffusion Adapter Using Controlled Multi-View Editing

Hansheng Chen, Ruoxi Shi, Yulin Liu, Bokui Shen, Jiayuan Gu, Gordon Wetzstein, Hao Su, Leonidas Guibas

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments V2 note: Fix missing acknowledgements. Project page: https://lakonik.github.io/mvedit

2311.17917 2023-11-30 cs.GR cs.CV 62%

AvatarStudio: High-fidelity and Animatable 3D Avatar Creation from Text

Jianfeng Zhang, Xuanmeng Zhang, Huichao Zhang, Jun Hao Liew, Chenxu Zhang, Yi Yang, Jiashi Feng

专题命中 3D生成：NeRF(abstract)；分类 cs.CV、cs.GR

Comments Project page at http://jeff95.me/projects/avatarstudio.html

2310.19415 2023-11-01 cs.CV cs.AI cs.GR 62%

Text-to-3D with Classifier Score Distillation

Xin Yu, Yuan-Chen Guo, Yangguang Li, Ding Liang, Song-Hai Zhang, Xiaojuan Qi

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments Our project page is https://xinyu-andy.github.io/Classifier-Score-Distillation

2308.14244 2023-08-29 cs.CV cs.GR 62%

HoloFusion: Towards Photo-realistic 3D Generative Modeling

Animesh Karnewar, Niloy J. Mitra, Andrea Vedaldi, David Novotny

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments ICCV 2023 conference; project page at: https://holodiffusion.github.io/holofusion

2304.12670 2023-04-27 cs.GR cs.CV 62%

Patch-based 3D Natural Scene Generation from a Single Example

Weiyu Li, Xuelin Chen, Jue Wang, Baoquan Chen

专题命中 3D生成：3D generation(abstract)；分类 cs.CV、cs.GR

Comments 23 pages, 26 figures, accepted by CVPR 2023. Project page: http://weiyuli.xyz/Sin3DGen/

1. 3D生成 447 篇

PartMat: Material-Aware 3D Part Decomposition with a Single Global Latent

SuperVoxelGPT: Adaptive and Ordered 3D Tokenization for Autoregressive Shape Generation

HAT-4D: Lifting Monocular Video for 4D Multi-Object Interactions via Human-Agent Collaboration

Arbor: Explicit Geometric Conditioning for Controllable 3D Asset Generation

AssetGen: Deployable 3D Asset Generation at Interactive Speed

WorldFlow3D: Flowing Through 3D Distributions for Unbounded World Generation

Recent Advances in 3D Object and Scene Generation: A Survey

PhysX-Anything: Simulation-Ready Physical 3D Assets from Single Image

FreeArt3D: Training-Free Articulated Object Generation using 3D Diffusion

Elevating 3D Models: High-Quality Texture and Geometry Refinement from a Low-Quality Model

MARVEL-40M+: Multi-Level Visual Elaboration for High-Fidelity Text-to-3D Content Creation

Kiss3DGen: Repurposing Image Diffusion Models for 3D Asset Generation

Generative AI Framework for 3D Object Generation in Augmented Reality

HIPPo: Harnessing Image-to-3D Priors for Model-free Zero-shot 6D Pose Estimation

BAG: Body-Aligned 3D Wearable Asset Generation

DI-PCG: Diffusion-based Efficient Inverse Procedural Content Generation for High-quality 3D Asset Creation

One Shot, One Talk: Whole-body Talking Avatar from a Single Image

DimensionX: Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion

Score Distillation via Reparametrized DDIM

3D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations

Exploiting Priors from 3D Diffusion Models for RGB-Based One-Shot View Planning

A Survey On Text-to-3D Contents Generation In The Wild

Coin3D: Controllable and Interactive 3D Assets Generation with Proxy-Guided Conditioning

GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models

SyncDreamer: Generating Multiview-consistent Images from a Single-view Image

Generic 3D Diffusion Adapter Using Controlled Multi-View Editing

AvatarStudio: High-fidelity and Animatable 3D Avatar Creation from Text

Text-to-3D with Classifier Score Distillation

HoloFusion: Towards Photo-realistic 3D Generative Modeling

Patch-based 3D Natural Scene Generation from a Single Example