Conference Programme

Noted that this program is tentative and subject to change

Programme booklet: CVMP 2023 Programme (9.6 MB)

Conference proceedings: ACM Digital Library

Thursday 30th November 2023
09:00 Registration opens
09:20 Welcome
Marco Volino, University of Surrey
09:30 Session #1: Image-based AI
  • HDR Illumination Outpainting with a Two-Stage GAN Model
    Jack Hilliard, Adrian Hilton, Jean-Yves Guillemaut
  • One-shot Detail Retouching with Patch Space Neural Transformation Blending
    Fazilet Gokbudak, Cengiz Oztireli
  • A Compact and Semantic Latent Space for Disentangled and Controllable Image Editing
    Gwilherm Lesné, Yann Gousseau, Saïd Ladjal, Alasdair Newson
  • DECORAIT - DECentralized Opt-in/out Registry for AI Training
    Kar Balan, Alex Black, Andrew Gilbert, Simon Jenni, Andy Parsons, John Collomosse
  • Expression-aware video inpainting for HMD removal in XR applications
    Fatemeh Ghorbani Lohesara, Karen Eguiazarian, Sebastian Knorr
11:10 Coffee Break
11:40 Keynote: The Magic of The Man in the Hat: Creating VFX for Indiana Jones and the Dial of Destiny
Andrew Whitehurst, Industrial Light and Magic
12:40 Spotlight Session-Short Papers and Demos
    Presented: 30th November
  • Light Field Video Compression Efficiency Through View Omission and Synthesis
    Nusrat Mehajabin (University of British Columbia); Tala Bazzaza (University of British Columbia); Hamid Reza Tohidypour (University of British Columbia)*; Yixiao Wang (University of British Columbia); Panos Nasiopoulos (University of British Columbia)
  • Human Pose Estimation Assisted Posture Reminder for E-Learning Video Production
    Yixia Zhao (Computer Network Information Center, Chinese Academy of Sciences)*; Zhenhua Feng (University of Surrey)
  • Practical Imaging and Analysis of Turbidity in Liquid
    Kewei Li (Imperial College London)*; Abhijeet Ghosh (Imperial College London)
  • SDiT: Enhanced Stable Diffusion for Audio-to-Talking Face Generation
    Fatemeh Nazarieh (University of Surrey)*; Zhenhua Feng (University of Surrey)
  • So you think you can dance? Controllable Multi-person Dance Generation
    Emily Corby (University of Surrey)*; Andrew Gilbert (University of Surrey); Edward Fish (University of Surrey)
  • InclusivityXR – an online tool for detecting inclusivity issues in AR and VR
    Andy T Woods (StoryFutures)*; Umar Farooq (University of Surrey); James Bennett (StoryFutures); Marco Volino (University of Surrey)
  • Presented: 1st December
  • Towards Neural Representations of Heterogeneous Translucent Voxelised Media
    Thomas Gillooly (Norwegian University of Science and Technology)*; Jon Yngve Hardeberg (NTNU); Abhijeet Ghosh (Imperial College London); Giuseppe Claudio Guarnera (University of York)
  • Refocus-NeRF: Focus-Distance-Aware Neural Radiance Fields Trained with Focus Bracket Photography
    Yuki Yabumoto (Shibaura Institute of Technology)*; Takuhiro Nishida (Shibaura Institute of Technology); Takashi Ijiri (Shibaura Institute of Technology)
  • MV-SyDog: A Multi-View 3D Dog Pose Dataset for Advancing 3D Pose Estimation
    Moira C Shooter (University of Surrey)*; Charles Malleson (University of Surrey); Adrian Hilton (University of Surrey)
  • AI Robotic Camera System for Live Soccer Broadcasts
    Kota Takahashi (Japan Broadcasting Corporation)*; Toshie Misu (Japan Broadcasting Corporation); Kensuke Hisatomi ( Japan Broadcasting Corporation)
  • Depth Reprojection for Mitigating Latency in XR Media Production
    Dominic Brown (disguise Technologies Ltd)*; Sebastian Day (disguise Technologies Ltd); Tom Whittock (disguise Technologies Ltd)
  • Real-time omnidirectional 3D multi-person human pose estimation with occlusion handling
    Pawel M Knap (University of Southampton)*; Peter TD Hardy (University of Southampton); Alberto Tamajo (University of Southampton); Hwasup Lim (Korea Institute of Science and Technology); Hansung Kim (University Of Southampton)
  • Demo Presentation
  • Using ML networks for turning 2D video into 3D Volumetric Video inside of Unreal Engine 5
    Alex Grona (Velox XR Limited)*
  • Drawing from Motion Capture: Performance, 3D Drawing and Character Animation
    Ricardo Megre (AIM Creative Studios)*
  • LightMoCap: Light-weight, Real-time & Scalable Markerless Motion Capture
    Georgios Albanis (University of Thessaly)*; NIKOLAOS ZIOULIS (Independent Researcher); Anargyros Chatzitofis (Moverse); Spyridon Thermos (Moverse); Vladimiros Sterzentsenko (CERTH); Kostas Kolomvatsos (University of Thessaly)
  • Audio-Driven Video Composition
    Tim Rumpf (Filmuniversity Babelsberg Konrad Wolf); Jakub Fiser (Adobe Research)*
13:00 Lunch, Posters and Demos
14:30 Session #2: Pose and Motion
  • Optimising 2D Pose Representations: Improving Accuracy, Stability and Generalisability Within Unsupervised 2D-3D Human Pose Estimation
    Peter Hardy, Srinandan Dasmahapatra, Hansung Kim
  • BundleMoCap: Efficient, Robust and Smooth Motion Capture from Sparse Multiview Videos
    Georgios Albanis, Nikolaos Zioulis, Kostas Kolomvatsos
  • CVMP Awards
    Jeff Clifford
15:30 Coffee Break
16:00 Keynote: Digital Humans in Generative Video
Jonathan Starck, Synthesia
17:00 Networking Reception
Friday 1st December 2023
09:00 Registration opens
09:30 Session #3: Image Processing and Reconstruction
  • Redistributing the Precision and Content in 3D-LUT-based Inverse Tone-mapping for HDR/WCG Display
    Cheng Guo, Leidong Fan, Qian Zhang, Hanyuan Liu, Kanglin Liu, Xiuhua Jiang
  • A software test bed for sharing and evaluating color transfer algorithms for images and 3D objects
    Herbert Potechius, Gunasekaran Raja, Thomas Sikora, Sebastian Knorr
  • LFSphereNet: Real Time Spherical Light Field Reconstruction from a Single Omnidirectional Image
    Manu Gond, Emin Zerman, Sebastian Knorr, Mårten Sjöström
  • View-dependent Adaptive HLOD: real-time interactive rendering of multi-resolution models
    Rui Li
10:50 Coffee Break
11:20Keynote: Towards Multi-Modal Generation
Duygu Ceylan, Adobe Research
12:20 Lunch, Posters and Demos
14:00 Session #4: Special Session GenAI and XR
  • Applications in Media for Novel View Synthesis
    Graeme Phillipson (BBC)*; Hell Raymond-Hayling (BBC); Alia Sheikh (BBC)
  • Making Gen AI work for us, rather than the other way around
    Will MacNeil (The Mill)
  • Overcoming Latency in Real-Time Virtual Production
    Dominic Brown (Disguise)
  • Beyond Reality - Towards Creative Applications for Mixed Reality
    Thu Nguyen-Phuoc (Reality Labs Research at Meta)
15:20 Coffee Break
15:50 Keynote: Engines For Everything Except Games
Tupac Martir, Satore Studio
16:50 Closing
Armin Mustafa, University of Surrey