3D 视觉 - arXivDaily 专题

2603.15433 2026-03-17 cs.CV 70%

Real-Time Human Frontal View Synthesis from a Single Image

从单张图像实时合成人体正面视图

Fangyu Lin, Yingdong Hu, Lunjie Zhu, Zhening Liu, Yushi Huang, Zehong Lin, Jun Zhang

机构 * HKUST（香港科技大学）

专题命中新视角合成：point cloud(abstract)；novel view synthesis(abstract)；分类 cs.CV

AI总结本文提出PrismMirror框架，通过级联学习策略实现单图像实时正面视图合成，提升视觉真实性和结构准确性，达到24 FPS的实时性能。

2603.12903 2026-03-16 cs.CV 70%

Spectral-Geometric Neural Fields for Pose-Free LiDAR View Synthesis

光谱-几何神经场用于无姿态激光雷达视角合成

Yinuo Jiang, Jun Cheng, Yiran Wang, Cheng Cheng

机构 * School of AIA, Huazhong University of Science and Technology（华中科技大学人工智能学院）

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

AI总结本文提出SG-NLF框架，通过融合光谱信息与几何一致性实现无姿态激光雷达视角合成，提升重建质量和姿态精度。

Comments Accepted by CVPR 2026

2603.09819 2026-03-11 cs.CV 70%

ConfCtrl: Enabling Precise Camera Control in Video Diffusion via Confidence-Aware Interpolation

ConfCtrl: 通过置信度感知插值实现视频扩散中的精确相机控制

Liudi Yang, George Eskandar, Fengyi Shen, Mohammad Altillawi, Yang Bai, Chi Zhang, Ziyuan Liu, Abhinav Valada

机构 * University of Freiburg（弗赖堡大学）； Ludwig Maximilian University of Munich（慕尼黑路德维希-马克西米利安大学）； Technical University of Munich（慕尼黑技术大学）； Huawei Heisenberg Research Center (Munich)（华为海森堡研究中心（慕尼黑））

专题命中新视角合成：point cloud(abstract)；novel view synthesis(abstract)；分类 cs.CV

AI总结 ConfCtrl通过置信度感知插值实现视频扩散中的精确相机控制，解决大视角变化下新视角生成问题，提升几何一致性与视觉合理性。

Comments 13 pages

2603.08147 2026-03-10 cs.CV 70%

MV-Fashion: Towards Enabling Virtual Try-On and Size Estimation with Multi-View Paired Data

MV-Fashion: 向通过多视角配对数据实现虚拟试衣和尺寸估计迈进

Hunor Laczkó, Libang Jia, Loc-Phat Truong, Diego Hernández, Sergio Escalera, Jordi Gonzalez, Meysam Madadi

机构 * Universitat Autònoma de Barcelona（巴塞罗那自治大学）； Computer Vision Center（计算机视觉中心）； Universitat de Barcelona（巴塞罗那大学）

专题命中新视角合成：point cloud(abstract)；novel view synthesis(abstract)；分类 cs.CV

AI总结 MV-Fashion通过多视角配对数据集提升虚拟试衣和服装尺寸估计任务的性能。

2602.16337 2026-02-19 cs.CV cs.LG 70%

Subtractive Modulative Network with Learnable Periodic Activations

减法调制网络与可学习周期激活

Tiou Wang, Zhuoqian Yang, Markus Flierl, Mathieu Salzmann, Sabine Süsstrunk

机构 * School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Sweden（电气工程与计算机科学学院，皇家理工学院，瑞典）； School of Computer and Communication Sciences, EPFL, Switzerland（计算机与通信科学学院，瑞士联邦理工学院）

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

AI总结 SMN通过可学习周期激活和调制掩码模块，实现了高效高保真图像重建和3D NeRF新视角合成。

Comments 4 pages, 3 figures, 3 tables

Journal ref ICASSP 2026-2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

2601.05722 2026-01-12 cs.CV 70%

Rotate Your Character: Revisiting Video Diffusion Models for High-Quality 3D Character Generation

旋转你的角色：重新审视视频扩散模型用于高质量3D角色生成

Jin Wang, Jianxiang Lu, Comi Chen, Guangzheng Xu, Haoyu Yang, Peng Chen, Na Zhang, Yifan Xu, Longhuang Wu, Shuai Shao, Qinglin Lu, Ping Luo

机构 * Hunyuan, Tencent（腾讯文予实验室）； The University of Hong Kong（香港大学）

专题命中新视角合成：novel view synthesis(abstract)；3D generation(abstract)；分类 cs.CV

AI总结本文提出RCM模型，通过旋转角色姿势实现高质量3D角色生成和新视角合成，支持多视角输入和高分辨率视频生成。

Comments 11 pages, 8 figures

2601.03362 2026-01-08 cs.CV 70%

Guardians of the Hair: Rescuing Soft Boundaries in Depth, Stereo, and Novel Views

毛发守护者：恢复深度、立体和新视角中的柔软边界

Xiang Zhang, Yang Zhang, Lukas Mehl, Markus Gross, Christopher Schroers

专题命中新视角合成：3D vision(abstract)；novel view synthesis(abstract)；分类 cs.CV

AI总结 HairGuard通过深度修复网络和生成场景画家恢复3D视觉中柔软边界的细节，提升单目深度估计、立体转换和新视角合成的性能。

2512.06818 2025-12-09 cs.CV 70%

MeshSplatting: Differentiable Rendering with Opaque Meshes

基于网格的可微渲染：使用不透明网格的可微渲染

Jan Held, Sanghyun Son, Renaud Vandeghen, Daniel Rebain, Matheus Gadelha, Yi Zhou, Anthony Cioppa, Ming C. Lin, Marc Van Droogenbroeck, Andrea Tagliasacchi

机构 * University of Liège（里耶克斯大学）； Simon Fraser University（西蒙弗雷泽大学）； University of Maryland（马里兰大学）； University of British Columbia（不列颠哥伦比亚大学）； University of Toronto（多伦多大学）； Adobe Research（Adobe研究院）

专题命中新视角合成：Gaussian Splatting(abstract)；novel view synthesis(abstract)；分类 cs.CV

AI总结 MeshSplatting通过可微渲染方法，实现了基于网格的高效重建，提升了实时3D场景交互的质量和性能。

2501.16312 2025-10-17 cs.CV 70%

LinPrim: Linear Primitives for Differentiable Volumetric Rendering

Nicolas von Lützow, Matthias Nießner

机构 * Technical University of Munich（慕尼黑技术大学）

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments Project page: https://nicolasvonluetzow.github.io/LinPrim - Project video: https://youtu.be/NRRlmFZj5KQ - Accepted at NeurIPS 2025

2510.12749 2025-10-15 cs.CV 70%

SPORTS: Simultaneous Panoptic Odometry, Rendering, Tracking and Segmentation for Urban Scenes Understanding

Zhiliu Yang, Jinyu Dai, Jianyuan Zhang, Zhu Yang

机构 * School of Information Science and Engineering, Yunnan University（云南大学信息科学与工程学院）； Yunnan University（云南大学）； School of Information and Electronics, Beijing Institute of Technology（北京理工大学信息与电子学院）； Yunnan Key Laboratory of Intelligent Systems and Computing（智能系统与计算重点实验室）

专题命中新视角合成：point cloud(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments Accepted by IEEE Transactions on Multimedia

2510.07190 2025-10-09 cs.CV 70%

MV-Performer: Taming Video Diffusion Model for Faithful and Synchronized Multi-view Performer Synthesis

Yihao Zhi, Chenghong Li, Hongjie Liao, Xihe Yang, Zhengwentai Sun, Jiahao Chang, Xiaodong Cun, Wensen Feng, Xiaoguang Han

机构 * Great Bay University（大亚湾大学）； School of Artificial Intelligence, Shenzhen University（深圳大学人工智能学院）； Guangdong Provincial Key Laboratory of Future Networks of Intelligence（广东省未来网络智能化重点实验室）

专题命中新视角合成：point cloud(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments Accepted by SIGGRAPH Asia 2025 conference track

2507.15748 2025-10-01 cs.CV 70%

CHROMA: Consistent Harmonization of Multi-View Appearance via Bilateral Grid Prediction

Jisu Shin, Richard Shaw, Seunghyun Shin, Zhensong Zhang, Hae-Gon Jeon, Eduardo Perez-Pellitero

机构 * Huawei Noah’s Ark Lab（华为诺亚实验室）； GIST AI Graduate School（GIST人工智能研究生院）； Yonsei University（延世大学）

专题命中新视角合成：3D reconstruction(abstract)；novel view synthesis(abstract)；分类 cs.CV

2506.23835 2025-09-23 cs.CV 70%

SCORP: Scene-Consistent Object Refinement via Proxy Generation and Tuning

Ziwei Chen, Ziling Liu, Zitong Huang, Mingqi Gao, Feng Zheng

机构 * Southern University of Science and Technology（南方科技大学）； University of Sheffield（谢菲尔德大学）； Spatialtemporal AI（时空人工智能）

专题命中新视角合成：novel view synthesis(abstract)；3D generation(abstract)；分类 cs.CV

Comments 8 pages with 6 figures. Project page: https://polysummit.github.io/scorp.github.io/

2509.10241 2025-09-16 cs.CV 70%

On the Geometric Accuracy of Implicit and Primitive-based Representations Derived from View Rendering Constraints

Elias De Smijter, Renaud Detry, Christophe De Vleeschouwer

机构 * KU Leuven, Dept. Electrical Engineering, Research unit Processing Speech and Images（KU莱顿，电气工程系，语音与图像研究单位）； UCLouvain, ICTEAM, Dept. Electrical Engineering（UCLouvain，ICTEAM，电气工程系）； KU Leuven, Dept. Mechanical Engineering, Research unit Robotics, Automation and Mechatronics（KU莱顿，机械工程系，机器人、自动化与机电研究单位）

专题命中新视角合成：Gaussian Splatting(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments 9 pages, 3 figures, to be presented at ASTRA25,

2404.00992 2025-09-10 cs.CV 70%

SGCNeRF: Few-Shot Neural Rendering via Sparse Geometric Consistency Guidance

Yuru Xiao, Xianming Liu, Deming Zhai, Kui Jiang, Junjun Jiang, Xiangyang Ji

机构 * School of Computer Science and Technology, Harbin Institute of Technology（计算机科学与技术学院，哈尔滨工业大学）； Department of Automation, Tsinghua University（自动化系，清华大学）

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments Accepted by IEEE Transactions on Circuits and Systems for Video Technology

2412.13188 2025-08-27 cs.CV 70%

StreetCrafter: Street View Synthesis with Controllable Video Diffusion Models

Yunzhi Yan, Zhen Xu, Haotong Lin, Haian Jin, Haoyu Guo, Yida Wang, Kun Zhan, Xianpeng Lang, Hujun Bao, Xiaowei Zhou, Sida Peng

机构 * Zhejiang University（浙江大学）； Li Auto Inc.（Li汽车公司）； Cornell University（康奈尔大学）

专题命中新视角合成：point cloud(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments Project page: https://zju3dv.github.io/street_crafter

2508.12615 2025-08-20 cs.CV 70%

WIPES: Wavelet-based Visual Primitives

Wenhao Zhang, Hao Zhu, Delong Wu, Di Kang, Linchao Bao, Xun Cao, Zhan Ma

机构 * Nanjing University（南京大学）； Tencent（腾讯）

专题命中新视角合成：3D vision(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments IEEE/CVF International Conference on Computer Vision 2025

2508.06392 2025-08-11 cs.CV 70%

FVGen: Accelerating Novel-View Synthesis with Adversarial Video Diffusion Distillation

Wenbin Teng, Gonglin Chen, Haiwei Chen, Yajie Zhao

专题命中新视角合成：3D reconstruction(abstract)；novel view synthesis(abstract)；分类 cs.CV

2506.23153 2025-07-01 cs.CV 70%

Dynamic View Synthesis from Small Camera Motion Videos

Huiqiang Sun, Xingyi Li, Juewen Peng, Liao Shen, Zhiguo Cao, Ke Xian, Guosheng Lin

机构 * Key Laboratory of Image Processing and Intelligent Control, Ministry of Education（图像处理与智能控制重点实验室）； School of Artificial Intelligence and Automation, Huazhong University of Science and Technology（人工智能与自动化学院）； School of Electronic Information and Communications, Huazhong University of Science and Technology（电子信息与通讯学院）； S-Lab, Nanyang Technological University (NTU)（南洋理工大学S实验室）

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments Accepted by TVCG

2505.19813 2025-05-27 cs.CV 70%

GoLF-NRT: Integrating Global Context and Local Geometry for Few-Shot View Synthesis

You Wang, Li Fang, Hao Zhu, Fei Hu, Long Ye, Zhan Ma

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments CVPR 2025

2505.05848 2025-05-19 cs.CV 70%

RefRef: A Synthetic Dataset and Benchmark for Reconstructing Refractive and Reflective Objects

Yue Yin, Enze Tao, Weijian Deng, Dylan Campbell

专题命中新视角合成：3D reconstruction(abstract)；novel view synthesis(abstract)；分类 cs.CV

2505.00702 2025-05-02 cs.CV 70%

RayZer: A Self-supervised Large View Synthesis Model

Hanwen Jiang, Hao Tan, Peng Wang, Haian Jin, Yue Zhao, Sai Bi, Kai Zhang, Fujun Luan, Kalyan Sunkavalli, Qixing Huang, Georgios Pavlakos

机构 * The University of Texas at Austin（德克萨斯大学奥斯汀分校）； Adobe Research（Adobe研究）

专题命中新视角合成：3D vision(abstract)；novel view synthesis(abstract)；分类 cs.CV

2408.16690 2025-04-30 cs.CV 70%

Generic Objects as Pose Probes for Few-shot View Synthesis

Zhirui Gao, Renjiao Yi, Chenyang Zhu, Ke Zhuang, Wei Chen, Kai Xu

机构 * School of Computer, National University of Defense Technology（计算机学院，国防科技大学）

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments Accepted by IEEE TCSVT 2025 Project page: https://zhirui-gao.github.io/PoseProbe.github.io/

2411.19224 2025-03-21 eess.IV cs.CV physics.med-ph 70%

Differentiable Voxel-based X-ray Rendering Improves Sparse-View 3D CBCT Reconstruction

Mohammadhossein Momeni, Vivek Gopalakrishnan, Neel Dey, Polina Golland, Sarah Frisken

专题命中新视角合成：3D reconstruction(abstract)；novel view synthesis(abstract)；分类 cs.CV

2407.17470 2025-03-03 cs.CV 70%

SV4D: Dynamic 3D Content Generation with Multi-Frame and Multi-View Consistency

Yiming Xie, Chun-Han Yao, Vikram Voleti, Huaizu Jiang, Varun Jampani

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments Project page: https://sv4d.github.io/

2407.20908 2025-02-17 cs.CV 70%

Dynamic Scene Understanding through Object-Centric Voxelization and Neural Rendering

Yanpeng Zhao, Yiwei Hao, Siyu Gao, Yunbo Wang, Xiaokang Yang

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments Accepted by TPAMI2025

2406.07828 2024-12-16 cs.CV 70%

Spatial Annealing for Efficient Few-shot Neural Rendering

Yuru Xiao, Deming Zhai, Wenbo Zhao, Kui Jiang, Junjun Jiang, Xianming Liu

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments AAAI 2025, code available at https://github.com/pulangk97/SANeRF

2411.15706 2024-11-26 cs.CV cs.LG 70%

Fixing the Perspective: A Critical Examination of Zero-1-to-3

Jack Yu, Xueying Jia, Charlie Sun, Prince Wang

专题命中新视角合成：novel view synthesis(abstract)；3D generation(abstract)；分类 cs.CV

2411.13631 2024-11-25 cs.CV 70%

Sparse Input View Synthesis: 3D Representations and Reliable Priors

Nagabhushan Somraj

专题命中新视角合成：NeRF(abstract)；novel view synthesis(abstract)；分类 cs.CV

Comments PhD Thesis of Nagabhushan S N, Dept of ECE, Indian Institute of Science (IISc); Advisor: Dr. Rajiv Soundararajan; Thesis Reviewers: Dr. Kaushik Mitra (IIT Madras), Dr. Aniket Bera (Purdue University); Submitted: May 2024; Accepted and Defended: Sep 2024; Abstract condensed, please check the PDF for full abstract

2411.14208 2024-11-22 cs.CV 70%

Novel View Extrapolation with Video Diffusion Priors

Kunhao Liu, Ling Shao, Shijian Lu

专题命中新视角合成：point cloud(abstract)；novel view synthesis(abstract)；分类 cs.CV

1. 新视角合成 734 篇

Real-Time Human Frontal View Synthesis from a Single Image

Spectral-Geometric Neural Fields for Pose-Free LiDAR View Synthesis

ConfCtrl: Enabling Precise Camera Control in Video Diffusion via Confidence-Aware Interpolation

MV-Fashion: Towards Enabling Virtual Try-On and Size Estimation with Multi-View Paired Data

Subtractive Modulative Network with Learnable Periodic Activations

Rotate Your Character: Revisiting Video Diffusion Models for High-Quality 3D Character Generation

Guardians of the Hair: Rescuing Soft Boundaries in Depth, Stereo, and Novel Views

MeshSplatting: Differentiable Rendering with Opaque Meshes

LinPrim: Linear Primitives for Differentiable Volumetric Rendering

SPORTS: Simultaneous Panoptic Odometry, Rendering, Tracking and Segmentation for Urban Scenes Understanding

MV-Performer: Taming Video Diffusion Model for Faithful and Synchronized Multi-view Performer Synthesis

CHROMA: Consistent Harmonization of Multi-View Appearance via Bilateral Grid Prediction

SCORP: Scene-Consistent Object Refinement via Proxy Generation and Tuning

On the Geometric Accuracy of Implicit and Primitive-based Representations Derived from View Rendering Constraints

SGCNeRF: Few-Shot Neural Rendering via Sparse Geometric Consistency Guidance

StreetCrafter: Street View Synthesis with Controllable Video Diffusion Models

WIPES: Wavelet-based Visual Primitives

FVGen: Accelerating Novel-View Synthesis with Adversarial Video Diffusion Distillation

Dynamic View Synthesis from Small Camera Motion Videos

GoLF-NRT: Integrating Global Context and Local Geometry for Few-Shot View Synthesis

RefRef: A Synthetic Dataset and Benchmark for Reconstructing Refractive and Reflective Objects

RayZer: A Self-supervised Large View Synthesis Model

Generic Objects as Pose Probes for Few-shot View Synthesis

Differentiable Voxel-based X-ray Rendering Improves Sparse-View 3D CBCT Reconstruction

SV4D: Dynamic 3D Content Generation with Multi-Frame and Multi-View Consistency

Dynamic Scene Understanding through Object-Centric Voxelization and Neural Rendering

Spatial Annealing for Efficient Few-shot Neural Rendering

Fixing the Perspective: A Critical Examination of Zero-1-to-3

Sparse Input View Synthesis: 3D Representations and Reliable Priors

Novel View Extrapolation with Video Diffusion Priors