Novel View Synthesis

Ask about this topic

Overview

Novel view synthesis studies how to render a scene from camera positions that were not observed, supporting applications such as 3D reconstruction, virtual reality, robotics, and medical imaging. Common approaches include neural radiance fields, 3D Gaussian Splatting, and feed-forward multi-view transformers; diffusion and video-generation models are increasingly used to fill unseen regions or handle large viewpoint changes. Methods are tested on benchmarks such as LLFF, Mip-NeRF 360, DTU, and RealEstate10K, with growing attention to sparse, unposed, dynamic, and multimodal inputs. A central challenge is balancing photorealistic completion of unseen content with geometrically consistent views, especially when extrapolating beyond the training cameras, while keeping models efficient and robust to real-world conditions.

Latest papers

CardsList

289 papers33 papers in the last four weeks, up 14% on the four weeks before. 0.2% of all new papers.

  1. Generating Roadside LiDAR Datasets from Vehicle-Side Datasets via Novel View Synthesis

    May 7, 2026Yuhan Xia, Runxin Zhao, Hanyang Zhuang +2Point Cloud CompletionIntelligent Transportation Systems

    School of Automation and Intelligent Sensing, Shanghai Jiao Tong University, Shanghai 200240, China; Key Laboratory of System Control and Information Processing, Ministry of Education of China, Shanghai 200240, China · Global College, Shanghai Jiao Tong University, Shanghai 200240, China

  2. HumanSplatHMR: Closing the Loop Between Human Mesh Recovery and Gaussian Splatting Avatar

    May 4, 2026Yeheng Zong, Pou-Chun Kung, Yike Pan +43D ReconstructionHuman Mesh Recovery

    University of Michigan

  3. Faster 3D Gaussian Splatting Convergence via Structure-Aware Densification

    Apr 30, 2026Linjie Lyu, Ayush Tewari, Jianchun Chen +23D Gaussian Splatting3DGS Acceleration

    Max-Planck-Institut für Informatik, Germany · Cambridge University, United Kingdom · Saarbrücken Research Center for Visual Computing, Interaction, and Artificial Intelligence (VIA), Germany

  4. Softmax-GS: Generalized Gaussians Learning When to Blend or Bound

    Apr 30, 2026Chen Ziwen, Peng Wang, Hao Tan +2Multi-View Consistency3D Gaussian Splatting

    Adobe Research · Tripo AI · Hillbot +1

  5. Difix3D-W: Distractor-Free Few-Shot 3D Gaussian Splatting in the Wild

    Apr 30, 2026Wongi Park, Jordan A. James, Myeongseok Nam +43D Gaussian SplattingDiffusion Model Guidance

    1Ajou Univerity · University of Texas at Arlington · 3GenGenAI +2

  6. Two-View Accumulation as the Primary Training Lever for Hybrid-Capture Gaussian Splatting: A Variance-Decomposition View of When Gradient Surgery Helps

    Apr 29, 2026Sungjun Cho3D Gaussian SplattingMulti-View Learning

    The Hong Kong University of Science and Technology

  7. 3D-LENS: A 3D Lifting-based Elevated Novel-view Synthesis method for Single-View Aerial-Ground Re-Identification

    Apr 29, 2026William Grolleau, Astrid Sabourin, Guillaume Lapouge +1Novel View SynthesisPerson Re-Identification

    Université Paris-Saclay, CEA, List, F-91120 Palaiseau, France · Sorbonne University, CNRS, Institute of Intelligent Systems and Robotics (ISIR), 4 Place Jussieu, 75005 Paris, France

  8. Multivariate Gaussian NeRF for Wide Field-of-View Ultrasound Reconstruction

    Apr 27, 2026Patris Valera, Magdalena Wysocki, Felix Duelmer +4Novel View SynthesisUltrasound Imaging

    Chair for Computer Aided Medical Procedures (CAMP), Technical University of Munich, Germany · Munich Center for Machine Learning (MCML), Munich, Germany · LUMA Vision, Beech Hill Road, Dublin, Ireland

  9. Light 'em Up: Enabling Few-Shot Low-Light 3D Gaussian Splatting with Multi-Scale Explicit Retinex Illumination Decoupling

    Apr 27, 2026YuHao Yin, Zongji Wang, Yuanben Zhang +3Multi-View Consistency3D Gaussian Splatting

    Aerospace Information Research Institute, Chinese Academy of Sciences · Key Laboratory of Target Cognition and Application Technology (TCAT) · University of Chinese Academy of Sciences +2

  10. Spatiotemporal Degradation-Aware 3D Gaussian Splatting for Realistic Underwater Scene Reconstruction

    Apr 26, 2026Shaohua Liu, Ning Gao, Zuoya Gu +33D Gaussian SplattingUnderwater Imaging

    School of Astronautics Beihang University Beijing, China · Shen Yuan Honors College, Beihang University · School of Artificial Intelligence Beihang University Beijing, China +3

  11. VistaBot: View-Robust Robot Manipulation via Spatiotemporal-Aware View Synthesis

    Apr 23, 2026Songen Gu, Yuhang Zheng, Weize Li +6Generalization in Robotic ManipulationNovel View Synthesis

    1Fudan University · 2TARS Robotics · 4NUS +1

  12. AnyRecon: Arbitrary-View 3D Reconstruction with Video Diffusion Model

    Apr 21, 2026Yutian Chen, Shi Guo, Renbiao Jin +7Video Diffusion Models3D Reconstruction

    Shanghai AI Lab · Shanghai Jiao Tong University · The University of Hong Kong +2

  13. UniGeo: Unifying Geometric Guidance for Camera-Controllable Image Editing via Video Models

    Apr 19, 2026Hong Jiang, Wensong Song, Zongxin Yang +2Multi-View ConsistencyDiffusion Model Guidance

    ReLER, CCAI, Zhejiang University · DBMI, HMS, Harvard University

  14. GlobalSplat: Efficient Feed-Forward 3D Gaussian Splatting via Global Scene Tokens

    Apr 16, 2026Roni Itkin, Noam Issachar, Yehonatan Keypur +33D Gaussian Splatting3DGS Acceleration

    The Hebrew University of Jerusalem · Westlake University

  15. Hybrid Latents: Geometry-Appearance-Aware Surfel Splatting

    Apr 16, 2026Neel Kelkar, Simon Niedermayr, Klaus Engel +13D Gaussian SplattingMulti-View 3D Reconstruction

    Technical University of Munich, Munich, Germany · Siemens Healthineers, Erlangen, Germany

  16. Dehaze-then-Splat: Generative Dehazing with Physics-Informed 3D Gaussian Splatting for Smoke-Free Novel View Synthesis

    Apr 15, 2026Boss Chen, Hanqing WangNovel View SynthesisMulti-View Consistency

    The Hong Kong University of Science and Technology (Guangzhou)

  17. RobotPan: A 360∘^\circ Surround-View Robotic Vision System for Embodied Perception

    Apr 15, 2026Jiahao Ma, Qiang Zhang, Peiran Liu +11Robotic PerceptionPanoramic 3D Reconstruction

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

    Apr 15, 2026Cheng-You Lu, Yi-Shan Hung, Wei-Ling Chi +6Novel View SynthesisNeural Radiance Fields

    University of Technology Sydney · University of Sydney · National Yang Ming Chiao Tung University

  19. MSGS: Multispectral 3D Gaussian Splatting

    Apr 14, 2026Iris Zheng, Guojun Tang, Alexander Doronin +23D Gaussian SplattingNovel View Synthesis

    Victoria University of Wellington

  20. ArtifactWorld: Scaling 3D Gaussian Splatting Artifact Restoration via Video Generation Models

    Apr 14, 2026Xinliang Wang, Yifeng Shi, Zhenyu Wu3D Gaussian SplattingNovel View Synthesis

    Ke Holdings Inc. Beijing, China

  21. 3DTV: A Feedforward Interpolation Network for Real-Time View Synthesis

    Apr 13, 2026Stefan Schulz, Fernando Edelstein, Hannah Dröge +2Novel View SynthesisMulti-View Geometry

    University of Bonn

  22. FreeScale: Scaling 3D Scenes via Certainty-Aware Free-View Generation

    Apr 12, 2026Chenhan Jiang, Yu Chen, Qingwen Zhang +4Novel View SynthesisSynthetic Data Augmentation

    Hong Kong University of Science and Technology · National Univeristy of Singapore · KTH Royal Institute of Technology +3

  23. FrameCrafter: Novel View Synthesis as Video Completion

    Apr 9, 2026Qi Wu, Khiem Vuong, Minsik Jeon +2Video Diffusion ModelsNovel View Synthesis

    Carnegie Mellon University

  24. Geo-EVS: Geometry-Conditioned Extrapolative View Synthesis for Autonomous Driving

    Apr 8, 2026Yatong Lan, Rongkui Tang, Lei HeNovel View SynthesisAutonomous Driving

    School of Vehicle and Mobility, Tsinghua University, Beijing, China · State Key Laboratory of Intelligent Green Vehicle and Mobility, Tsinghua University, Beijing, China · School of Science, Minzu University, Beijing, China +1

  25. AnchorSplat: Feed-Forward 3D Gaussian Splatting with 3D Geometric Priors

    Apr 8, 2026Xiaoxue Zhang, Xiaoxu Zheng, Yixuan Yin +53D Gaussian Splatting3D Reconstruction

  26. LiveStre4m: Feed-Forward Live Streaming of Novel Views from Unposed Multi-View Video

    Apr 8, 2026Pedro Quesado, Erkut Akdag, Yasaman Kashefbahrami +2Novel View SynthesisDynamic Novel View Synthesis

    AIMSGroup, Department of Electrical Engineering, Eindhoven University of Technology

  27. 3D Smoke Scene Reconstruction Guided by Vision Priors from Multimodal Large Language Models

    Apr 7, 2026Xinye Zheng, Fei Wang, Yiqi Nie +53D ReconstructionMulti-View 3D Reconstruction

    Hefei University of Technology · Institute of Artificial Intelligence, Hefei Comprehensive National Science Center · Anhui University +2

  28. LSRM: High-Fidelity Object-Centric Reconstruction via Scaled Context Windows

    Apr 6, 2026Zhengqin Li, Cheng Zhang, Jakob Engel +1Object-Centric 3D ReconstructionInverse Rendering

    Meta Reality Labs Research

  29. GenSmoke-GS: A Multi-Stage Method for Novel View Synthesis from Smoke-Degraded Images Using a Generative Model

    Apr 3, 2026Qida Cao, Xinyuan Hu, Changyue Shi +3Novel View SynthesisMulti-View 3D Reconstruction

    School of Computer Science and Technology, Hangzhou Dianzi University · School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen

  30. NavCrafter: Exploring 3D Scenes from a Single Image

    Apr 3, 2026Hongbo Duan, Peiyu Zhuang, Yi Liu +5Novel View SynthesisSingle-View 3D Reconstruction

    Center for Artificial Intelligence and Robotics, Shenzhen International Graduate School, Tsinghua University, Shenzhen 518055, China · Peng Cheng Laboratory, 518108, China · School of Cyber Science and Technology, Shenzhen Campus of Sun Yat-sen University, China +1