Third Workshop for Learning 3D with Multi-View Supervision (3DMV)

at CVPR 2026




Call for papers:   February 1st, 2026

Submission Deadline:   March 4th March 6th, 2026, 11:59 PM GMT

Workshop Day:   June 4th, 2026, 8:30 AM

Location:   Room 703, Colorado Convention Center, Denver, Colorado, USA

Event Pictures

Third Workshop for Learning 3D with Multi-View Supervision @ CVPR 2026

The third 3DMV workshop at CVPR 2026 explores the rapidly expanding frontier of 3D and 4D research from multi-view data. Building on the success of the first and second editions that focused on 3D and object generation, 3DMV 2026 highlights dynamic multi-view datasets, generative 4D models, and the emerging paradigm of world models—systems that learn to predict and simulate evolving 3D scenes with spatial and temporal consistency. We also explore their integration with frontier techniques such as MLLMs and video diffusion. A special emphasis is placed on emerging applications in robotics and medical AI, where 3D/4D world models enable robust generalization. 3DMV 2026 aims to unify 2D, 3D, and 4D research communities and foster emerging applications toward scalable, cross-domain multi-view learning. The detailed topics covered in the workshop include the following:

  • Multi-View for 3D Understanding
  • Deep Multi-View Stereo
  • Multi-View for 3D Generation and Novel View Synthesis
  • Dynamic Multi-View Datasets and 4D Generative models
  • MLLMs and World Models for 3D/4D
  • Video Diffusion for Multi-View Generation
  • Robotics Applications with Multi-View 3D/4D
  • Medical AI with Multi-View 3D/4D

    • Submission TimeLine

  • Paper Submission start: February 1st, 2026
  • Paper submission deadline: March 4th March 6th, 2026, 11:59 PM GMT
  • Review period: March 6th - March 18th, 2026
  • Decision to authors: March 23rd, 2026
  • Camera-ready papers: April 2nd, 2026

    • Call For Papers

    We are soliciting papers that use Multi-view deep learning to address problems in 3D/4D Understanding and Generation, including but not limited to the following topics:

  • Bird-Eye View for 3D Object Detection
  • Multi-view fusion for 3D Object Detection
  • Indoor/outdoor scenes segmentation
  • 3D/4D Diffusions for generation
  • Video diffusion for multi-view synthesis
  • 4D understanding, generation, and world models
  • MLLMs and foundation models for 3D/4D
  • Language + 3D/4D
  • Medical 3D segmentation and analysis
  • Robotics with multi-view 3D/4D perception
  • 3D shape generation and reconstruction
  • Deep multi-view stereo
  • Inverse Graphics from multi-view images
  • Indoor/outdoor scenes generation and reconstruction
  • Volumetric Multi-view representation for 3D generation and novel view synthesis
  • NeRFs and Gaussian Splatting
  • 3D shape classification and retrieval
  • Vision for XR, AR, VR

    • Paper Submission Guidelines

  • We accept both archival and non-archival paper submissions.
  • Archival submissions should be of max 8 pages (excluding references) on the aforementioned and related topics.
  • Non-archival submissions can be previously published works in major venues (in the last two years or at CVPR 2026) or based on new works (max 8 pages as well).
  • Archival Accepted papers will be included in the proceedings of CVPR 2026, while non-archival accpeted papers will not be included
  • Submitted manuscripts should follow the CVPR 2026 paper template (if they have not been published previously).
  • All submissions (except for previoulsly published) will be peer-reviewed under a double-blind policy (authors should not include names in submissions)
  • PDFs need to be submitted online through the CMT link.
  • Accepted papers' authors will be notified to prepare camera-ready posters to be uploaded based on the schedule above.
  • Every accepted paper will have the opportunity to host a poster presentation at the workshop.
  • There will be a `best poster award` announced during the workshop with a sponsored money prize.

    • Schedule

    June 4th, 2026, 8:30 AM · Room 703, Colorado Convention Center, Denver, Colorado, USA

    Time Session Speakers
    8:30 – 8:45 Opening Remarks Guocheng Qian
    8:45 - 9:25 3D Scene Reconstruction Matthias Niessner
    9:25 - 10:05 Multi-view Generative Diffusion Models Ziwei Liu
    10:05 - 11:05 Coffee Break & Posters Session
    11:05 - 11:45 Building a 3D Foundation for Spatial AI Andrea Vedaldi
    11:45 - 12:15 Closing & Panel Discussion All speakers and moderators
    12:15 - 13:00 Lunch Break

    Poster session logistics (10:05–11:05 AM @ ExHall A): 16 assigned boards (IDs 10–25) for 31 posters, allocated first come, first serve — please try to fit two posters per board where possible. Once those fill up, use any free board in the 306–352 range (available before 11:30 AM). Please remove your poster at the end of the session.

    Speakers

    Matthias Niessner

    Matthias Niessner
    Technical University of Munich
    Ziwei Liu

    Ziwei Liu
    Nanyang Technological University
    Andrea Vedaldi

    Andrea Vedaldi
    University of Oxford

    Paper Awards

    Congratulations to the recipients of the 3DMV @ CVPR 2026 Best Paper Award and Best Paper Runner-up Award.

    DanceNet3D: A 3D Dance Dataset with Multi-View Videos and 3DGS ReconstructionsBest Paper
    Shihang Wei (New York University); Mingjian Li (New York University); Ran Gong (New York University); Yueyu Hu (New York University); Yao Wang (New York University)
    PDF
    TerraSky3D: Multi-View Reconstructions of European Landmarks in 4KRunner-up
    Mattia D'Urso (TU Graz); Yuxi Hu (TU Graz); Christian Sormann (Sony); Mattia Rossi (Sony); Friedrich Fraundorfer (TU Graz)
    PDF

    Accepted Papers

    We are pleased to announce the 24 papers accepted to the Third Workshop for Learning 3D with Multi-View Supervision (3DMV) at CVPR 2026. Proceedings links will be added here as they become available; in the meantime, camera-ready PDFs are linked below where authors have provided them.

    CoGS-SLAM: Correspondence-Guided Gaussian Splatting SLAM with One-Shot Dense InitializationCamera-ready
    Wei Tse Cheng (National Yang Ming Chiao Tung University)
    PDF
    Lotus-2: Advancing Geometric Dense Prediction with Powerful Image Generative ModelCamera-ready
    Jing He (HKUST(GZ)); Haodong Li (HKUST(GZ)); Mingzhi Sheng (HKUST(GZ)); Yingcong Chen (HKUST(GZ))
    PDF
    PePESeg3D: Perception Prior Enhances Multi-Scale Segmentation for 3D Gaussian SplattingCamera-ready
    Sungjae Choi (KAIST); Seunghee Ko (KAIST); Junmo Kim (KAIST)
    PDF
    Surf-NeRF: Surface Regularised Neural Radiance Fields
    Jack Naylor (The University of Sydney); Viorela Ila (The University of Sydney); Donald Dansereau (The University of Sydney)
    PDF
    TerraSky3D: Multi-View Reconstructions of European Landmarks in 4K
    Mattia D'Urso (TU Graz); Yuxi Hu (TU Graz); Christian Sormann (Sony); Mattia Rossi (Sony); Friedrich Fraundorfer (TU Graz)
    PDF
    Towards Minimal Focal Stack in Shape from Focus
    Khurram Ashfaq (Korea University of Technology and Education); Muhammad Tariq Mahmood (Korea University of Technology and Education)
    PDF
    CylinderDepth: Cylindrical Spatial Attention for Multi-View Consistent Self-Supervised Surround Depth Estimation
    Samer Abualhanud (Lebniz University Hannover); Christian Grannemann (Lebniz University Hannover); Max Mehltretter (Lebniz University Hannover)
    PDF
    Benchmarking Stereo Geometry Estimation in the Wild
    Jeff Tan (Carnegie Mellon University); Nikhil Keetha (Carnegie Mellon University); Yifei Liu (Carnegie Mellon University); Shubham Tulsiani (Carnegie Mellon University); Deva Ramanan (Carnegie Mellon University)
    PDF
    LiveStre4m: Feed-Forward Live Streaming of Novel Views from Unposed Multi-View Video
    Pedro Quesado dos Santos (Eindhoven University of Technology); Erkut Akdag (Eindhoven University of Technology); Yasaman Kashefbahrami (Eindhoven University of Technology); Willem Menu (Eindhoven University of Technology); Egor Bondarev (Eindhoven University of Technology)
    PDF
    In Depth We Trust: Reliable Monocular Depth Supervision for Gaussian Splatting
    Wenhui Xiao (Queensland University of Technology); Ethan Goan (Queensland University of Technology); Rodrigo Santa Cruz (Queensland University of Technology); David Ahmedt-Aristizabal (CSIRO); Olivier Salvado (Queensland University of Technology); Clinton Fookes (Queensland University of Technology); Leo Lebrat (Queensland University of Technology)
    PDF
    PointSplat: Efficient Geometry-Driven Pruning and Transformer Refinement for 3D Gaussian Splatting
    Anh Thuan Tran (George Mason University); Jana Kosecka (George Mason University)
    PDF
    SDFoam: Signed-Distance Foam for Explicit Surface Reconstruction
    Antonella Rech (University of Trento); Nicola Conci (University of Trento); Nicola Garau (University of Trento)
    PDF
    Indoor Asset Detection in Large Scale 360° Drone-Captured Imagery via 3D Gaussian Splatting
    Monica Tang (UC Berkeley); Avideh Zakhor (UC Berkeley)
    PDF
    SymmFusion: Injecting Multi-View-Consistent Generative Priors into Symmetry-Guided Point Cloud Completion
    Bhanu Paregi (IISER Bhopal); Vaibhav Kumar (IISER Bhopal)
    PDF
    Real-time Gaussian Splatting with Progressive Block-based 3D Reconstruction
    Pirazh Khorramshahi (Qualcomm Technologies); Upal Mahbub (Qualcomm Technologies); Mehul Arora (Qualcomm Technologies); Adithya Nallabolu (Meta); Gokce Dane (Qualcomm Technologies)
    PDF
    Reducing Closeup Frequency Artifacts for Level-of-Detail 3D Gaussian Splatting
    Leonardo Zhou (University of Waterloo); Dayou Mao (University of Waterloo); Yuchen Lin (University of Waterloo); Ashkan Ebadi (National Research Council of Canada); Alexander Wong (University of Waterloo); Yuhao Chen (University of Waterloo)
    PDF
    Leveraging Mamba and Attention for Robust Multi-View Pedestrian Tracking
    Reef Alturki (University of Surrey); Faegheh Sardari (Microsoft); Adrian Hilton (University of Surrey); Jean-Yves Guillemaut (University of Surrey)
    PDF
    SplatTree: Voxel-Based Multi-Resolution Decoding for Feed-Forward 3D Gaussian Generation
    Ran Gong (New York University); Zhehong Ren (New York University); Tingyu Fan (New York University); Yueyu Hu (New York University); Yao Wang (New York University)
    PDF
    Scene-Agnostic Object-Centric Representation Learning for 3D Gaussian Splatting
    Tsuheng Hsu (Aalto University); Guiyu Liu (University of Oulu); Juho Kannala (Aalto University); Janne Heikkila (University of Oulu)
    PDF
    DanceNet3D: A 3D Dance Dataset with Multi-View Videos and 3DGS Reconstructions
    Shihang Wei (New York University); Mingjian Li (New York University); Ran Gong (New York University); Yueyu Hu (New York University); Yao Wang (New York University)
    PDF
    NGOcc: Normal-Guided 3D Occupancy Prediction with Observability-aware Time Fusion
    Jufei Wang (University of Science and Technology of China); Mingwei Wei (University of Science and Technology of China); Tianheng Qiu (University of Science and Technology of China); Hu Wei (Hefei Institutes of Physical Science, Chinese Academy of Sciences); Huasheng Ni (Hefei Institutes of Physical Science, Chinese Academy of Sciences); Xuan Huang (Hefei Institutes of Physical Science, Chinese Academy of Sciences)
    PDF
    F3G-Avatar: Face Focussed Full-body Gaussian Avatar
    Willem Menu (Technical University of Eindhoven)
    PDF
    Coordinate-Reshaped Triplane Video Representation for Dynamic Gaussian Splatting Compression
    Dae Yeol Lee (Dolby Laboratories); Birendra Kathariya (Dolby Laboratories); Tsung-Wei Huang (Dolby Laboratories); Guan-Ming Su (Dolby Laboratories); Peng Yin (Dolby Laboratories)
    PDF
    PAGaS: Pixel-Aligned 1DoF Gaussian Splatting for Depth Refinement
    David Recansens (Universidad de Zaragoza); Robert Maier; Aljaz Bozic (Meta); Stephane Grabli (Meta); Javier Civera (Universidad de Zaragoza); Tony Tung; Edmond Boyer (INRIA)
    PDF

    Organizers

    Abdullah Hamdi
    KAUST

    Guocheng Qian
    Snap Research

    Jinjie Mai
    KAUST

    Sara Rojas Martinez
    KAUST

    Jan Held
    University of Liege

    Silvio Giancola
    KAUST

    Program Committee (Reviewers)

    Akhmedkhan Shabanov (Simon Fraser University)  ·  Boris Chidlovskii (Naver Labs Europe)  ·  Daniel Duckworth (Google DeepMind)  ·  Gopal Sharma (UBC)  ·  Gordon Guocheng Qian (Snap)  ·  Jan Held (University of Liège)  ·  Jiading Fang (Toyota Technological Institute at Chicago)  ·  Jianqi Chen (Beihang University)  ·  Jinjie Mai (KAUST)  ·  Nitish Agarwal (KinaTrax)  ·  Priyam Parashar (FAIR, Meta)  ·  Renaud Vandeghen (University of Liège)  ·  Sara Rojas Martinez (KAUST)  ·  Shuzhou Yang (Peking University)  ·  Silvio Giancola (KAUST)  ·  Tak Yeon Lee (KAIST)  ·  Tianhao Wu (Nanyang Technological University)  ·  Vitor Guizilini (Toyota Research Institute)  ·  Xingguang Yan (Simon Fraser University)  ·  Yan Xia (Technical University of Munich)

    Contact: [email protected]
    CVPR 2026 Workshop ©2026
    The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.