3D 视觉 - arXivDaily 专题

2402.14586 2024-02-27 cs.CV cs.GR 84%

FrameNeRF: A Simple and Efficient Framework for Few-shot Novel View Synthesis

Yan Xing, Pan Wang, Ligang Liu, Daolun Li, Li Zhang

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

2312.09743 2023-12-18 cs.CV cs.GR 84%

SLS4D: Sparse Latent Space for 4D Novel View Synthesis

Qi-Yuan Feng, Hao-Xiang Chen, Qun-Ce Xu, Tai-Jiang Mu

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

Comments 10 pages, 6 figures

2311.04634 2023-11-09 cs.CV cs.GR 84%

VET: Visual Error Tomography for Point Cloud Completion and High-Quality Neural Rendering

Linus Franke, Darius Rückert, Laura Fink, Matthias Innmann, Marc Stamminger

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

2207.05736 2022-10-17 cs.CV cs.GR 84%

Vision Transformer for NeRF-Based View Synthesis from a Single Input Image

Kai-En Lin, Lin Yen-Chen, Wei-Sheng Lai, Tsung-Yi Lin, Yi-Chang Shih, Ravi Ramamoorthi

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

Comments WACV 2023 Project website: https://cseweb.ucsd.edu/~viscomp/projects/VisionNeRF/

2109.01018 2021-09-03 cs.GR cs.CV 84%

Dynamic Scene Novel View Synthesis via Deferred Spatio-temporal Consistency

Beatrix-Emőke Fülöp-Balogh, Eleanor Tursman, James Tompkin, Julie Digne, Nicolas Bonneel

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

Comments Accompanying video: https://youtu.be/RXK2iv980nU

2604.13416 2026-07-07 cs.CV cs.AI 版本更新 83%

DF3DV-1K: A Large-Scale Dataset and Benchmark for Distractor-Free Novel View Synthesis

DF3DV-1K：用于无干扰新视角合成的大规模数据集与基准

Cheng-You Lu, Yi-Shan Hung, Wei-Ling Chi, Hao-Ping Wang, Charlie Li-Ting Tsai, Yu-Cheng Chang, Yu-Lun Liu, Thomas Do, Chin-Teng Lin

机构 * University of Technology Sydney（悉尼科技大学）； University of Sydney（悉尼大学）； National Yang Ming Chiao Tung University（阳明交通大学）

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

AI总结为弥补无干扰辐射场领域缺乏大规模真实世界数据集的空白，构建了包含1048个场景、每场景提供干净和杂乱图像集的DF3DV-1K数据集，并基于此基准测试了九种最新方法，识别出最鲁棒的方法和最具挑战的场景。

Comments ECCV 2026 Accepted

2606.23027 2026-06-26 cs.CV 新提交 83%

Learning Stable Canonical Worlds for Novel View Synthesis and Beyond

学习稳定规范世界用于新视角合成及更多应用

Xiaoyu Xu, Jian Zou, Sheyang Tang, Zhihua Wang, Jing Liao, Kede Ma

机构 * City University of Hong Kong（香港城市大学）； University of Waterloo（滑铁卢大学）

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

AI总结提出CanonicalGS，一种前馈高斯泼溅管道，通过不确定性感知融合将多视图观测映射到稳定场景表示，提升新视角合成质量并迁移到下游视觉任务。

2506.10981 2026-06-17 cs.CV 版本更新 83%

SceneCompleter: Dense 3D Scene Completion for Generative Novel View Synthesis

SceneCompleter：面向生成式新视角合成的密集3D场景补全

Weiliang Chen, Jiayi Bi, Yuanhui Huang, Wenzhao Zheng, Yueqi Duan

机构 * Department of Electronic Engineering, Tsinghua University（电子工程系，清华大学）； Department of Automation, Tsinghua University（自动化系，清华大学）

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

AI总结提出SceneCompleter，通过几何-外观双流扩散模型在RGBD潜空间进行密集3D场景补全，并引入场景嵌入器整合全局信息，实现跨视角一致的生成式新视角合成。

2603.20176 2026-06-02 cs.CV 83%

LagerNVS: Latent Geometry for Fully Neural Real-time Novel View Synthesis

LagerNVS：用于全神经实时新视角合成的潜在几何

Stanislaw Szymanowicz, Minghao Chen, Jianyuan Wang, Christian Rupprecht, Andrea Vedaldi

机构 * Visual Geometry Group, University of Oxford（牛津大学视觉几何组）； Meta AI

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

AI总结提出LagerNVS，一种基于3D感知潜在特征的编码器-解码器神经网络，通过显式3D监督预训练初始化编码器，结合轻量解码器和光度损失端到端训练，实现实时、泛化的新视角合成，在Re10k上达到31.4 PSNR。

Comments IEEE CVF Conference on Computer Vision and Pattern Recognition 2026. Project page with code, models and examples: szymanowiczs.github.io/lagernvs

2605.07287 2026-05-22 cs.CV 83%

SplatWeaver: Learning to Allocate Gaussian Primitives for Generalizable Novel View Synthesis

SplatWeaver: 学习分配高斯原语以实现可泛化的新型视角合成

Yecong Wan, Fan Li, Mingwen Shao, Wangmeng Zuo

机构 * Faculty of Computing, Harbin Institute of Technology（哈尔滨工业大学计算机学院）； Zhengzhou Advanced Research Institute of Harbin Institute of Technology（哈尔滨工业大学郑州先进研究院）； Huawei Noah’s Ark Lab（华为诺亚实验室）； Artificial Intelligence Research Institute, Shenzhen University of Advanced Technology（深圳先进技术大学人工智能研究院）

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

AI总结本文提出SplatWeaver框架，通过动态分配高斯原语实现可泛化的新型视角合成，解决传统方法中固定分配导致的资源浪费和表达不足问题。

Comments Project Page: https://yecongwan.github.io/SplatWeaver/

2603.20428 2026-03-24 cs.CV 83%

Benchmarking Efficient & Effective Camera Pose Estimation Strategies for Novel View Synthesis

对新颖视角合成中高效且有效的相机姿态估计策略进行基准测试

Jhacson Meza, Martin R. Oswald, Torsten Sattler

机构 * Faculty of Electrical Engineering, Czech Technical University in Prague（布拉格捷克技术大学电子工程系）； Czech Institute of Informatics, Robotics and Cybernetics, Czech Technical University in Prague（捷克技术大学信息技术、机器人与自动化研究所）； University of Amsterdam（阿姆斯特丹大学）

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

AI总结本文提出一个基准测试，评估在新颖视角合成中高效且有效的相机姿态估计策略，通过减少特征数量和使用前馈网络结合经典SfM方法，实现效率与精度的平衡。

2509.21853 2026-01-29 cs.CV 83%

Dynamic Novel View Synthesis in High Dynamic Range

高动态范围动态新视角合成

Kaixuan Zhang, Zhipeng Xiong, Minxian Li, Mingwu Ren, Jiankang Deng, Xiatian Zhu

机构 * Nanjing University of Science and Technology（南京理工大学）； Imperial College London（帝国理工学院）； University of Surrey（萨里大学）

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

AI总结本文提出HDR-4DGS方法，通过动态色调映射模块实现高动态范围动态新视角合成，提升时间辐射一致性和空间颜色准确性。

Comments It has been accepted by ICLR 2026

2512.22706 2025-12-30 cs.CV 83%

SCPainter: A Unified Framework for Realistic 3D Asset Insertion and Novel View Synthesis

SCPainter：一个用于真实3D资产插入和新视角合成的统一框架

Paul Dobre, Jackson Cooper, Xin Wang, Hongzhou Yang

机构 * Department of Geomatics Engineering, University of Calgary, Calgary, Alberta, Canada（测绘工程系，卡尔加里大学，卡尔加里，阿尔伯塔，加拿大）

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

AI总结 SCPainter通过整合3D高斯点和点云与扩散生成，实现真实3D资产插入与新视角合成，提升自动驾驶训练数据的多样性和真实性。

2501.02913 2025-12-25 cs.CV 83%

Pointmap-Conditioned Diffusion for Consistent Novel View Synthesis

基于点图的扩散模型用于一致的新型视角合成

Thang-Anh-Quan Nguyen, Nathan Piasco, Luis Roldão, Moussab Bennehar, Dzmitry Tsishkou, Laurent Caraffa, Jean-Philippe Tarel, Roland Brémond

机构 * Noah’s Ark, Huawei Paris Research Center（Noah’s Ark，华为巴黎研究中心）； COSYS, Gustave Eiffel University（COSYS，格拉夫-埃菲尔大学）； LASTIG, IGN-ENSG, Gustave Eiffel University（LASTIG，IGN-ENSG，格拉夫-埃菲尔大学）

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

AI总结 PointmapDiff通过利用点图作为条件信号，结合预训练的2D扩散模型，实现城市驾驶场景中一致的新型视角合成，并能生成高质量结果。

Comments WACV 2026. Project page: https://ntaquan0125.github.io/pointmap-conditioned-diffusion

2512.13796 2025-12-17 cs.CV 83%

Nexels: Neurally-Textured Surfels for Real-Time Novel View Synthesis with Sparse Geometries

Nexels：基于神经纹理的 surfels 用于具有稀疏几何的实时新视角合成

Victor Rong, Jan Held, Victor Chu, Daniel Rebain, Marc Van Droogenbroeck, Kiriakos N. Kutulakos, Andrea Tagliasacchi, David B. Lindell

机构 * University of Toronto（多伦多大学）； Vector Institute（向量研究所）； Simon Frasier University（西蒙弗雷泽大学）； University of Liège（列日大学）； University of British Columbia（不列颠哥伦比亚大学）

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

AI总结 Nexels 通过结合神经场和 surfels 实现高效的实时新视角合成，使用更少的原始体和内存，同时保持高质量的视觉效果。

Comments Webpage at https://lessvrong.com/cs/nexels

2512.13639 2025-12-16 cs.CV 83%

Charge: A Comprehensive Novel View Synthesis Benchmark and Dataset to Bind Them All

Charge：一个综合的新型视图合成基准和数据集，用于绑定所有内容

Michal Nazarczuk, Thomas Tanay, Arthur Moreau, Zhensong Zhang, Eduardo Pérez-Pellitero

机构 * Huawei Noah’s Ark Lab（华为诺亚实验室）

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

AI总结 Charge数据集通过高保真RGB图像和多种互补模态，为视图合成和3D视觉研究提供丰富的基准测试资源。

Comments Project page: https://charge-benchmark.github.io/

2511.13121 2025-11-18 cs.CV 83%

CloseUpShot: Close-up Novel View Synthesis from Sparse-views via Point-conditioned Diffusion Model

Yuqi Zhang, Guanying Chen, Jiaxing Chen, Chuanyu Fu, Chuan Huang, Shuguang Cui

机构 * Shenzhen Future Network of Intelligence Institute (FNii-Shenzhen)（深圳未来网络智能研究院（FNii-深圳））； Chinese University of Hong Kong at Shenzhen (CUHKSZ)（香港中文大学（深圳））； Sun Yat-sen University（中山大学）

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

Comments Project Link: https://zyqz97.github.io/CloseUpShot/

2510.06694 2025-10-09 cs.CV 83%

SCas4D: Structural Cascaded Optimization for Boosting Persistent 4D Novel View Synthesis

Jipeng Lyu, Jiahua Dong, Yu-Xiong Wang

机构 * University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）

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

Comments Published in Transactions on Machine Learning Research (06/2025)

2409.10925 2025-09-09 cs.CV 83%

HGSLoc: 3DGS-based Heuristic Camera Pose Refinement

Zhongyan Niu, Zhen Tan, Jinpu Zhang, Xueliang Yang, Dewen Hu

机构 * the College of Intelligence Science and Technology, National University of Defense Technology（智能科学与技术学院，国防科技大学）

专题命中新视角合成：3DGS(title)；NeRF(abstract)；3D reconstruction(abstract)；分类 cs.CV

Journal ref Proc. IEEE International Conference on Robotics and Automation (ICRA), 2025

2506.05558 2025-06-09 cs.CV 83%

On-the-fly Reconstruction for Large-Scale Novel View Synthesis from Unposed Images

Andreas Meuleman, Ishaan Shah, Alexandre Lanvin, Bernhard Kerbl, George Drettakis

机构 * Inria, Université Côte d’Azur France（法国国家信息与自动化技术研究院、阿尔卑斯大学）； TU Wien Austria（维也纳技术大学）

专题命中新视角合成：novel view synthesis(title)；Gaussian Splatting(abstract)；3DGS(abstract)；分类 cs.CV

Journal ref ACM Transactions on Graphics 44, 4 (August 2025)

2402.02906 2025-06-06 cs.CV cs.LG 83%

ViewFusion: Learning Composable Diffusion Models for Novel View Synthesis

Bernard Spiegl, Andrea Perin, Stéphane Deny, Alexander Ilin

机构 * Aalto University（阿alto大学）； System 2 AI

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

Comments Version accepted to TMLR

Journal ref Transactions on Machine Learning Research, May 2025

2505.13440 2025-05-20 cs.CV 83%

Recollection from Pensieve: Novel View Synthesis via Learning from Uncalibrated Videos

Ruoyu Wang, Yi Ma, Shenghua Gao

机构 * The University of Hong Kong（香港大学）

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

Comments 13 pages, 4 figures

2410.01425 2025-03-13 cs.CV 83%

EVA-Gaussian: 3D Gaussian-based Real-time Human Novel View Synthesis under Diverse Multi-view Camera Settings

Yingdong Hu, Zhening Liu, Jiawei Shao, Zehong Lin, Jun Zhang

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

2405.15518 2024-10-01 cs.CV 83%

Feature Splatting for Better Novel View Synthesis with Low Overlap

T. Berriel Martins, Javier Civera

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

2409.02048 2024-09-04 cs.CV 83%

ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesis

Wangbo Yu, Jinbo Xing, Li Yuan, Wenbo Hu, Xiaoyu Li, Zhipeng Huang, Xiangjun Gao, Tien-Tsin Wong, Ying Shan, Yonghong Tian

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

Comments Project page: https://drexubery.github.io/ViewCrafter/

2407.10482 2024-07-16 cs.CV 83%

NGP-RT: Fusing Multi-Level Hash Features with Lightweight Attention for Real-Time Novel View Synthesis

Yubin Hu, Xiaoyang Guo, Yang Xiao, Jingwei Huang, Yong-Jin Liu

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

Comments ECCV 2024

2403.10953 2024-06-25 cs.CV 83%

Ctrl123: Consistent Novel View Synthesis via Closed-Loop Transcription

Hongxiang Zhao, Xili Dai, Jianan Wang, Shengbang Tong, Jingyuan Zhang, Weida Wang, Lei Zhang, Yi Ma

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

2312.04551 2024-04-02 cs.CV 83%

Free3D: Consistent Novel View Synthesis without 3D Representation

Chuanxia Zheng, Andrea Vedaldi

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

Comments webpage: https://chuanxiaz.com/free3d/, code: https://github.com/lyndonzheng/Free3D

2312.08892 2023-12-15 cs.CV 83%

VaLID: Variable-Length Input Diffusion for Novel View Synthesis

Shijie Li, Farhad G. Zanjani, Haitam Ben Yahia, Yuki M. Asano, Juergen Gall, Amirhossein Habibian

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

Comments paper and supplementary material

2309.17390 2023-10-02 cs.CV 83%

Forward Flow for Novel View Synthesis of Dynamic Scenes

Xiang Guo, Jiadai Sun, Yuchao Dai, Guanying Chen, Xiaoqing Ye, Xiao Tan, Errui Ding, Yumeng Zhang, Jingdong Wang

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

Comments Accepted by ICCV2023 as oral. Project page: https://npucvr.github.io/ForwardFlowDNeRF

Journal ref ICCV2023

1. 新视角合成 732 篇

FrameNeRF: A Simple and Efficient Framework for Few-shot Novel View Synthesis

SLS4D: Sparse Latent Space for 4D Novel View Synthesis

VET: Visual Error Tomography for Point Cloud Completion and High-Quality Neural Rendering

Vision Transformer for NeRF-Based View Synthesis from a Single Input Image

Dynamic Scene Novel View Synthesis via Deferred Spatio-temporal Consistency

DF3DV-1K: A Large-Scale Dataset and Benchmark for Distractor-Free Novel View Synthesis

Learning Stable Canonical Worlds for Novel View Synthesis and Beyond

SceneCompleter: Dense 3D Scene Completion for Generative Novel View Synthesis

LagerNVS: Latent Geometry for Fully Neural Real-time Novel View Synthesis

SplatWeaver: Learning to Allocate Gaussian Primitives for Generalizable Novel View Synthesis

Benchmarking Efficient & Effective Camera Pose Estimation Strategies for Novel View Synthesis

Dynamic Novel View Synthesis in High Dynamic Range

SCPainter: A Unified Framework for Realistic 3D Asset Insertion and Novel View Synthesis

Pointmap-Conditioned Diffusion for Consistent Novel View Synthesis

Nexels: Neurally-Textured Surfels for Real-Time Novel View Synthesis with Sparse Geometries

Charge: A Comprehensive Novel View Synthesis Benchmark and Dataset to Bind Them All

CloseUpShot: Close-up Novel View Synthesis from Sparse-views via Point-conditioned Diffusion Model

SCas4D: Structural Cascaded Optimization for Boosting Persistent 4D Novel View Synthesis

HGSLoc: 3DGS-based Heuristic Camera Pose Refinement

On-the-fly Reconstruction for Large-Scale Novel View Synthesis from Unposed Images

ViewFusion: Learning Composable Diffusion Models for Novel View Synthesis

Recollection from Pensieve: Novel View Synthesis via Learning from Uncalibrated Videos

EVA-Gaussian: 3D Gaussian-based Real-time Human Novel View Synthesis under Diverse Multi-view Camera Settings

Feature Splatting for Better Novel View Synthesis with Low Overlap

ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesis

NGP-RT: Fusing Multi-Level Hash Features with Lightweight Attention for Real-Time Novel View Synthesis

Ctrl123: Consistent Novel View Synthesis via Closed-Loop Transcription

Free3D: Consistent Novel View Synthesis without 3D Representation

VaLID: Variable-Length Input Diffusion for Novel View Synthesis

Forward Flow for Novel View Synthesis of Dynamic Scenes