3D 视觉 - arXivDaily 专题

2110.14373 2021-10-28 cs.CV cs.GR cs.LG 62%

Neural-PIL: Neural Pre-Integrated Lighting for Reflectance Decomposition

Mark Boss, Varun Jampani, Raphael Braun, Ce Liu, Jonathan T. Barron, Hendrik P. A. Lensch

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

Comments Project page: https://markboss.me/publication/2021-neural-pil/ Video: https://youtu.be/AsdAR5u3vQ8 - Accepted at NeurIPS 2021

2106.14942 2021-10-28 cs.CV cs.GR cs.LG 62%

Fast Training of Neural Lumigraph Representations using Meta Learning

Alexander W. Bergman, Petr Kellnhofer, Gordon Wetzstein

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

Comments Project website: http://www.computationalimaging.org/publications/metanlr/

2110.12993 2021-10-26 cs.CV cs.GR cs.LG 62%

Neural Relightable Participating Media Rendering

Quan Zheng, Gurprit Singh, Hans-Peter Seidel

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

Comments Accepted to NeurIPS 2021

2109.09378 2021-09-21 cs.GR cs.CV 62%

FreeStyleGAN: Free-view Editable Portrait Rendering with the Camera Manifold

Thomas Leimkühler, George Drettakis

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

Comments Project webpage: https://repo-sam.inria.fr/fungraph/freestylegan/

Journal ref ACM Transactions on Graphics (SIGGRAPH Asia 2021)

2012.09854 2021-08-19 cs.CV cs.AI cs.GR cs.LG stat.ML 62%

Worldsheet: Wrapping the World in a 3D Sheet for View Synthesis from a Single Image

Ronghang Hu, Nikhila Ravi, Alexander C. Berg, Deepak Pathak

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

Comments ICCV 2021; 17 pages

2108.05577 2021-08-13 cs.CV cs.GR 62%

iButter: Neural Interactive Bullet Time Generator for Human Free-viewpoint Rendering

Liao Wang, Ziyu Wang, Pei Lin, Yuheng Jiang, Xin Suo, Minye Wu, Lan Xu, Jingyi Yu

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

Comments Accepted by ACM MM 2021

2001.04947 2021-07-07 cs.GR cs.CV 62%

Neural Human Video Rendering by Learning Dynamic Textures and Rendering-to-Video Translation

Lingjie Liu, Weipeng Xu, Marc Habermann, Michael Zollhoefer, Florian Bernard, Hyeongwoo Kim, Wenping Wang, Christian Theobalt

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

2104.10786 2021-04-23 cs.CV cs.RO 62%

Exploring 2D Data Augmentation for 3D Monocular Object Detection

Sugirtha T, Sridevi M, Khailash Santhakumar, B Ravi Kiran, Thomas Gauthier, Senthil Yogamani

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

2103.05606 2021-04-13 cs.CV cs.GR cs.LG 62%

NeX: Real-time View Synthesis with Neural Basis Expansion

Suttisak Wizadwongsa, Pakkapon Phongthawee, Jiraphon Yenphraphai, Supasorn Suwajanakorn

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

Comments CVPR 2021 (Oral)

2103.11571 2021-03-23 cs.CV cs.GR 62%

Neural Lumigraph Rendering

Petr Kellnhofer, Lars Jebe, Andrew Jones, Ryan Spicer, Kari Pulli, Gordon Wetzstein

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

Comments Project website: http://www.computationalimaging.org/publications/nlr/

2103.06432 2021-03-12 cs.CV cs.RO 62%

Robust 2D/3D Vehicle Parsing in CVIS

Hui Miao, Feixiang Lu, Zongdai Liu, Liangjun Zhang, Dinesh Manocha, Bin Zhou

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

2008.01815 2020-08-06 cs.CV cs.GR 62%

Deep Multi Depth Panoramas for View Synthesis

Kai-En Lin, Zexiang Xu, Ben Mildenhall, Pratul P. Srinivasan, Yannick Hold-Geoffroy, Stephen DiVerdi, Qi Sun, Kalyan Sunkavalli, Ravi Ramamoorthi

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

Comments Published at the European Conference on Computer Vision, 2020

2004.03805 2020-04-09 cs.CV cs.GR 62%

State of the Art on Neural Rendering

Ayush Tewari, Ohad Fried, Justus Thies, Vincent Sitzmann, Stephen Lombardi, Kalyan Sunkavalli, Ricardo Martin-Brualla, Tomas Simon, Jason Saragih, Matthias Nießner, Rohit Pandey, Sean Fanello, Gordon Wetzstein, Jun-Yan Zhu, Christian Theobalt, Maneesh Agrawala, Eli Shechtman, Dan B Goldman, Michael Zollhöfer

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

Comments Eurographics 2020 survey paper

1904.05124 2020-04-08 cs.CV cs.AI cs.GR 62%

Predicting Novel Views Using Generative Adversarial Query Network

Phong Nguyen-Ha, Lam Huynh, Esa Rahtu, Janne Heikkila

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

Comments 12 pages, 4 figures, accepted for presentation at the Scandinavian Conference on Image Analysis 2019

1811.04756 2019-07-02 cs.GR cs.CV 62%

Deep-learning the Latent Space of Light Transport

Pedro Hermosilla, Sebastian Maisch, Tobias Ritschel, Timo Ropinski

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

Comments Eurographics Symposium on Rendering 2019

1903.07614 2019-05-07 cs.GR cs.CV cs.DS physics.data-an physics.geo-ph 62%

HexaShrink, an exact scalable framework for hexahedral meshes with attributes and discontinuities: multiresolution rendering and storage of geoscience models

Jean-Luc Peyrot, Laurent Duval, Frédéric Payan, Lauriane Bouard, Lénaïc Chizat, Sébastien Schneider, Marc Antonini

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

1904.12356 2019-04-30 cs.CV cs.GR 62%

Deferred Neural Rendering: Image Synthesis using Neural Textures

Justus Thies, Michael Zollhöfer, Matthias Nießner

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

Comments Video: https://youtu.be/z-pVip6WeyY SIGGRAPH 2019

1901.06034 2019-01-21 cs.CV cs.GR eess.IV 62%

High-speed Video from Asynchronous Camera Array

Si Lu

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

Comments 10 pages, 82 figures, Published at IEEE WACV 2019

2310.08587 2024-02-21 cs.CV 61%

Pseudo-Generalized Dynamic View Synthesis from a Video

Xiaoming Zhao, Alex Colburn, Fangchang Ma, Miguel Angel Bautista, Joshua M. Susskind, Alexander G. Schwing

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

Comments ICLR 2024; Originally titled as "Is Generalized Dynamic Novel View Synthesis from Monocular Videos Possible Today?"; Project page: https://xiaoming-zhao.github.io/projects/pgdvs

2603.05868 2026-08-11 cs.RO 版本更新 57%

AnyCamVLA: Zero-Shot Camera Adaptation for Viewpoint Robust Vision-Language-Action Models

AnyCamVLA: 零样本相机适应用于视角鲁棒的视觉-语言-动作模型

Hyeongjun Heo, Seungyeon Woo, Sang Min Kim, Junho Kim, Junho Lee, Yonghyeon Lee, Young Min Kim

机构 * Department of Electrical and Computer Engineering, Seoul National University（电子与计算机工程系，首尔国立大学）； Department of Mechanical Engineering, Massachusetts Institute of Technology（机械工程系，麻省理工学院）

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

AI总结 AnyCamVLA通过零样本相机适应提升视觉-语言-动作模型在视角变化下的鲁棒性，适用于任何RGB策略。

Comments Accepted to IROS 2026

2511.20853 2026-08-06 cs.CV cs.AI cs.LG eess.IV 版本更新 57%

MODEST: Multi-Optics Depth-of-Field Stereo Dataset

MODEST：多光学深度景深立体数据集

Nisarg K. Trivedi, Vinayaka A. Belludi, Li-Yun Wang

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

AI总结本文提出首个高分辨率立体DSLR数据集，涵盖18000张图像，系统变化焦距和光圈，用于研究单目和立体深度估计及景深渲染等任务。

Comments Website, dataset and software tools now available for purely non-commercial, academic research purposes. Significant new contributions. Project page: https://modest-dataset.netlify.app/ Huggingface: https://huggingface.co/datasets/independent-vision-lab/MODEST

2606.28828 2026-07-28 cs.CV 版本更新 57%

Ground4D: Consistency-Aware 4D Reconstruction from Monocular Video

Ground4D: 从单目视频进行一致性感知的四维重建

Qing Zhao, Weijian Deng, Pengxu Wei, Liang Lin

机构 * School of Computer Science and Engineering, Sun Yat-sen University, China（中山大学计算机科学与工程学院）； Tsinghua Shenzhen International Graduate School, Tsinghua University, China（清华大学深圳国际研究生院）

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

AI总结提出Ground4D框架，结合3D基础模型进行几何初始化与动态高斯泼溅进行一致性优化，实现从单目视频的高保真4D重建与新视角渲染。

2607.07320 2026-07-09 cs.CV 新提交 57%

SoccerNet 2026 Challenges Results

SoccerNet 2026挑战结果

Anthony Cioppa, Silvio Giancola, Håkan Ardö, Mohamad Dalal, Jan Held, Jérémie Ochin, Jiayuan Rao, Karen Sanchez, Renaud Vandeghen, Artur Xarles, Olivier Barnich, Albert Clapés, Mathieu Delvaux, Sergio Escalera, Bernard Ghanem, Cédric Hons, Antoine Houet, Sotiris Manitsaris, Tom Michel, Pierre Miralles, Thomas B. Moeslund, Mikael Nilsson, Bogdan Stanciulescu, Marc Van Droogenbroeck, Yanfeng Wang, Weidi Xie, Faisal Altawijri, Mohamed Atef, Semen Budennyy, Vasiliy Chelpanov, Puhua Chen, Yixin Chen, Lechao Cheng, Jianling Chu, Ju-Seong Do, Oleg Durygin, Omar Fetouh, Mirco Fuchs, Youssef Ghallab, Falguni Ghosh, Wonjun Heo, Yufeng Hu, Weixuan Huang, Phuong-Linh Huynh-Ha, Matvey Isupov, Yangguang Ji, Siyuan Jiang, Zhenxiang Jiang, Wonyong Jo, Ho-Young Jung, SeongHeon Kang, MinJae Kim, Youngseon Kim, Jakub Komosa, Artem Konshin, Trung-Hoang Le, Jongmin Lee, Lingling Li, Litao Li, Vadim Linkov, Fang Liu, Haoxuan Ma, Shun Makino, Ismail Mathkour, Konstantin Mitin, Mikhail Moiseev, Takumi Nagaya, Yuki Nakamura, Thanh-Khoi Nguyen, Hoang-Phuc Nguyen, Trong-Thuan Nguyen, Christian Orduz, Kwanyong Park, Fabian Perez, Parthsarthi Rawat, SuHyun Rim, Hoover Rueda-Chacón, Atom Scott, Minori Sugimura, Yuyang Sun, Shengeng Tang, Minh-Triet Tran, Ikuma Uchida, Juan Vanegas, Thanh-Nhan Vo, Jiangtao Wang, Yaxiong Wang, Xiaogang Wang, Ruifeng Wang, Rio Watanabe, Jiali Wen, Yongliang Wu, Di Yang, Xu Yang, Zhuo Yang, Xinyu Ye, Yibo Yu, Zihan Zhai, Yu Zhang, Zhenyu Zhao, Zhun Zhong, Yixi Zhou, Xingyu Zhu, Wenbo Zhu, Julian Ziegler

机构 * University of Liège（列日大学）； King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）； Spiideo（斯皮迪奥公司）； Aalborg University（奥尔堡大学）； SpAItial（斯帕蒂亚尔公司）； Center for Robotics, Mines Paris, PSL（巴黎矿业学院机器人中心（巴黎文理研究大学））； Footovision（Footovision公司）； Shanghai Jiao Tong University（上海交通大学）； Universitat de Barcelona（巴塞罗那大学）； Computer Vision Center（计算机视觉中心）； EVS Broadcast Equipment（EVS广播设备公司）； Pioneer Center for Artificial Intelligence（先锋人工智能中心）； Lund University（隆德大学）； TAHAKOM（TAHAKOM公司）； Mohamed Bin Zayed University for Artificial Intelligence (MBZUAI)（穆罕默德·本·扎耶德人工智能大学）； Sber AI（Sber人工智能公司）； Salute For Business（Salute For Business公司）； Intelligent Perception and Image Understanding Lab, Xidian University（西安电子科技大学智能感知与图像理解实验室）； South China University of Technology（华南理工大学）； Hefei University of Technology（合肥工业大学）； Kyungpook National University（庆北国立大学）； Leipzig University of Applied Sciences（莱比锡应用科学大学）； Friedrich-Alexander University Erlangen-Nuremberg（埃尔朗根-纽伦堡大学）； University of Seoul（首尔大学）； Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China（电子科技大学深圳高等研究院）； Nanjing University（南京大学）； University of Science, Ho Chi Minh City（胡志明市科技大学）； Nanyang Technological University（南洋理工大学）； National University of Singapore（新加坡国立大学）

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

AI总结 SoccerNet 2026挑战涵盖五项体育视频理解视觉任务，为每项任务提供数据、协议和基线。众多团队参与，本文介绍任务、评估协议，展示排行榜并总结领先提交内容，记录各任务当前状态。

Comments 40 pages

2607.02712 2026-07-07 cs.CV 新提交 57%

Global Pose Control for Generative View Synthesis in Normalized Object Coordinate Space

归一化物体坐标空间中用于生成视图合成的全局姿态控制

Zhibing Li, Amogh Gupta, Behnoosh Parsa, Dan Casas

机构 * The Chinese University of Hong Kong（香港中文大学）； Amazon（亚马逊）

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

AI总结针对现有生成模型在目标视图全局控制上的不足，提出在归一化物体坐标空间中精确相机控制的新方法，以单张或少量无姿态图像为输入，将视图合成视为图像编辑问题，提升了视图生成质量和保真度。

Comments Accepted to ECCV 2026. Project page https://lizb6626.github.io/GlobalNVS/

2512.05672 2026-07-07 cs.CV cs.AI cs.LG 版本更新 57%

InverseCrafter: Efficient Video ReCapture as a Latent Domain Inverse Problem

InverseCrafter：作为潜在域逆问题的高效视频重新捕捉

Yeobin Hong, Suhyeon Lee, Hyungjin Chung, Jong Chul Ye

机构 * Graduate School of AI, KAIST, South Korea（韩国釜山国立大学人工智能研究生院）； EverEx, South Korea（韩国EverEx公司）； Korea University, South Korea（韩国国立庆熙大学）

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

AI总结提出InverseCrafter框架，将新视角视频生成转化为潜在空间基于图像修复的逆问题，通过轻量级潜在掩码编码器建立算子等价，无需标注4D训练数据，实现高效合成与编辑。

Comments ECCV 2026

2512.20907 2026-07-02 cs.CV 版本更新 57%

PanoGrounder: Bridging 2D and 3D with Panoramic Scene Representations for VLM-based 3D Visual Grounding

PanoGrounder: 利用全景场景表示桥接2D和3D，实现基于VLM的3D视觉定位

Seongmin Jung, Seongho Choi, Gunwoo Jeon, Minsu Cho, Jongwoo Lim

机构 * Seoul National University（首尔大学）； Robotics Lab, Hyundai Motor Company（现代汽车公司机器人实验室）； Pohang University of Science and Technology (POSTECH)（浦项科技大学）

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

AI总结提出PanoGrounder框架，通过多模态全景表示与预训练2D VLM结合，实现强泛化能力的3D视觉定位，在ScanRefer和Nr3D上取得最优结果。

Comments ECCV 2026

2606.31585 2026-07-01 cs.CV cs.AI 新提交 57%

DPPE: Rethinking Camera-Based Positional Encoding for Scaling Multi-View Transformers

DPPE：重新思考基于相机的可缩放多视图Transformer的位置编码

Shun Kenney, Teppei Suzuki

机构 * Keio University（庆应义塾大学）； SB Intuitions

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

AI总结针对多视图Transformer中基于相机参数的位置编码导致训练后期性能停滞的问题，提出解耦位姿位置编码（DPPE），将旋转和平移显式分离，实现稳定长训练并提升外推泛化能力。

2604.18747 2026-07-01 cs.CV 版本更新 57%

URoPE: Universal Relative Position Embedding across Geometric Spaces

URoPE: 在几何空间中实现通用的相对位置嵌入

Yichen Xie, Depu Meng, Chensheng Peng, Yihan Hu, Quentin Herau, Masayoshi Tomizuka, Wei Zhan

机构 * Applied Intuition（应用直觉）； University of California, Berkeley（加州大学伯克利分校）

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

AI总结 URoPE扩展了RoPE，使模型能在不同视角或维度的几何空间中处理位置信息，适用于多任务如视图合成、3D检测等，提升模型在几何推理中的表现。

Comments Accepted by ECCV 2026. Code is available: https://urope-pe.github.io/

2604.26520 2026-06-30 cs.CV 57%

3D-LENS: A 3D Lifting-based Elevated Novel-view Synthesis method for Single-View Aerial-Ground Re-Identification

3D-LENS：一种基于3D提升的单视角空地重识别方法

William Grolleau, Astrid Sabourin, Guillaume Lapouge, Catherine Achard

机构 * Université Paris-Saclay, CEA, List（巴黎萨克雷大学、CEA、List）； Sorbonne University, CNRS, Institute of Intelligent Systems and Robotics (ISIR)（索邦大学、CNRS、智能系统与机器人研究所）

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

AI总结本文提出3D-LENS方法，通过结合几何一致的新型视角合成与鲁棒表示学习，解决单视角空地重识别中的视角域差距问题，实现跨视角检索。

Comments 15 pages, 2 figures, accepted to the European Conference on Computer Vision (ECCV) 2026

2606.27575 2026-06-29 cs.CV 新提交 57%

Perceptual 3D Simulation With Physical World Modeling

基于物理世界建模的感知3D仿真

Wanhee Lee, Klemen Kotar, Rahul Mysore Venkatesh, Jared Watrous, Daniel L. K. Yamins

机构 * Stanford University（斯坦福大学）

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

AI总结提出P3Sim系统，结合学习型物理世界模型、几何条件模块和持久场景记忆，在部分观测和不完整3D变换信号下模拟未来场景状态，实现多任务泛化。

Comments Published as a conference paper at CVPR 2026

1. 新视角合成 733 篇

Neural-PIL: Neural Pre-Integrated Lighting for Reflectance Decomposition

Fast Training of Neural Lumigraph Representations using Meta Learning

Neural Relightable Participating Media Rendering

FreeStyleGAN: Free-view Editable Portrait Rendering with the Camera Manifold

Worldsheet: Wrapping the World in a 3D Sheet for View Synthesis from a Single Image

iButter: Neural Interactive Bullet Time Generator for Human Free-viewpoint Rendering

Neural Human Video Rendering by Learning Dynamic Textures and Rendering-to-Video Translation

Exploring 2D Data Augmentation for 3D Monocular Object Detection

NeX: Real-time View Synthesis with Neural Basis Expansion

Neural Lumigraph Rendering

Robust 2D/3D Vehicle Parsing in CVIS

Deep Multi Depth Panoramas for View Synthesis

State of the Art on Neural Rendering

Predicting Novel Views Using Generative Adversarial Query Network

Deep-learning the Latent Space of Light Transport

HexaShrink, an exact scalable framework for hexahedral meshes with attributes and discontinuities: multiresolution rendering and storage of geoscience models

Deferred Neural Rendering: Image Synthesis using Neural Textures

High-speed Video from Asynchronous Camera Array

Pseudo-Generalized Dynamic View Synthesis from a Video

AnyCamVLA: Zero-Shot Camera Adaptation for Viewpoint Robust Vision-Language-Action Models

MODEST: Multi-Optics Depth-of-Field Stereo Dataset

Ground4D: Consistency-Aware 4D Reconstruction from Monocular Video

SoccerNet 2026 Challenges Results

Global Pose Control for Generative View Synthesis in Normalized Object Coordinate Space

InverseCrafter: Efficient Video ReCapture as a Latent Domain Inverse Problem

PanoGrounder: Bridging 2D and 3D with Panoramic Scene Representations for VLM-based 3D Visual Grounding

DPPE: Rethinking Camera-Based Positional Encoding for Scaling Multi-View Transformers

URoPE: Universal Relative Position Embedding across Geometric Spaces

3D-LENS: A 3D Lifting-based Elevated Novel-view Synthesis method for Single-View Aerial-Ground Re-Identification

Perceptual 3D Simulation With Physical World Modeling