Peter Hedman
- Senior Research Scientist, Google DeepMind (London) [firstname].j.[lastname]@gmail.com Twitter Google ScholarHi! I'm Peter Hedman, a researcher at Google DeepMind in London.
My focus is on creating immersive 3D experiences from easy-to-capture footage of real places. For example, you may want to scan and digitally revisit your childhood home, or capture a VR-ready 3D panorama of the places you visit during a vacation.
Formally my research interests are view synthesis, image-based rendering, neural rendering and real-time graphics.
PUBLICATIONS
EVER: Exact Volumetric Ellipsoid Rendering for Real-time View Synthesis (half-potato.gitlab.io/posts/ever) Alexander Mai, Peter Hedman, George Kopanas, Dor Verbin, David Futschik, Qiangeng Xu, Falko Kuester, Jonathan T. Barron, Yinda Zhang arXiv 2024 |
|
NeRF-Casting: Improved View-Dependent Appearance with Consistent Reflections (nerf-casting.github.io) Dor Verbin, Pratul Srinivasan, Peter Hedman, Ben Mildenhall, Benjamin Attal, Richard Szeliski, Jonathan T. Barron SIGGRAPH Asia 2024 |
|
Flash Cache: Reducing Bias in Radiance Cache Based Inverse Rendering (benattal.github.io/flash-cache) Benjamin Attal, Dor Verbin, Ben Mildenhall, Peter Hedman, Jonathan T. Barron, Matt O'Toole, Pratul Srinivasan ECCV 2024 (Oral) |
|
Binary Opacity Grids: Capturing Fine Geometric Detail for Mesh-Based View Synthesis (creiser.github.io/binary_opacity_grid) Christian Reiser, Stephan Garbin, Pratul Srinivasan, Dor Verbin, Richard Szeliski, Ben Mildenhall, Jonathan T. Barron, Peter Hedman*, Andreas Geiger* SIGGRAPH 2024 |
|
SMERF: Streamable Memory Efficient Radiance Fields for Real-Time Large-Scene Exploration (smerf-3d.github.io) Daniel Duckworth*, Peter Hedman*, Christian Reiser, Peter Zhizhin, Jean-François Thibert, Mario Lučić, Richard Szeliski, Jonathan T. Barron SIGGRAPH 2024 (Honorable Mention) |
|
InterNeRF: Scaling Radiance Fields via Parameter Interpolation (arxiv.org/abs/2406.11737) Clinton Wang, Peter Hedman, Polina Golland, Jonathan T. Barron, Daniel Duckworth CVPR Neural Rendering Intelligence 2024 |
|
Eclipse: Disambiguating Illumination and Materials using Unintended Shadows (dorverbin.github.io/eclipse) Dor Verbin, Ben Mildenhall, Peter Hedman, Jonathan T. Barron, Todd Zickler, Pratul Srinivasan CVPR 2024 (Oral) |
|
Inpaint3D: 3D Scene Content Generation using 2D Inpainting Diffusion (inpaint3d.github.io) Kira Prabhu*, Jane Wu*, Lynn Tsai*, Peter Hedman, Dan B Goldman, Ben Poole, Michael Broxton arXiv 2023 |
|
Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fields (jonbarron.info/zipnerf) Jonathan T. Barron, Ben Mildenhall, Dor Verbin, Pratul Srinivasan, Peter Hedman ICCV 2023 (Oral, Best Paper Finalist) |
|
Vox-E: Text-guided Voxel Editing of 3D Objects (tau-vailab.github.io/Vox-E) Etai Sella, Gal Fiebelman, Peter Hedman, Hadar Averbuch-Elor ICCV 2023 |
|
MERF: Memory-Efficient Radiance Fields for Real-time View Synthesis in Unbounded Scenes (creiser.github.io/merf) Christian Reiser, Richard Szeliski, Dor Verbin, Pratul Srinivasan, Ben Mildenhall, Andreas Geiger, Jonathan T. Barron, Peter Hedman SIGGRAPH 2023 |
|
BakedSDF: Meshing Neural SDFs for Real-Time View Synthesis (bakedsdf.github.io) Lior Yariv*, Peter Hedman*, Christian Reiser, Dor Verbin, Pratul Srinivasan, Richard Szeliski, Jonathan T. Barron, Ben Mildenhall SIGGRAPH 2023 |
|
MobileNeRF: Exploiting the Polygon Rasterization Pipeline for Efficient Neural Field Rendering on Mobile Architectures (mobile-nerf.github.io) Zhiqin Chen, Thomas Funkhouser, Peter Hedman, Andrea Tagliasacchi CVPR 2023 (Best Paper Award Candidate) |
|
AligNeRF: High-Fidelity Neural Radiance Fields via Alignment-Aware Training (yifanjiang19.github.io/alignerf) Yifan Jiang, Peter Hedman, Ben Mildenhall, Dejia Xu, Jonathan T. Barron, Zhangyang Wang, Tianfan Xue CVPR 2023 |
|
NeRF in the Dark: High Dynamic Range View Synthesis from Noisy Raw Images (bmild.github.io/rawnerf) Ben Mildenhall, Peter Hedman, Ricardo Martin-Brualla, Pratul Srinivasan, Jonathan T. Barron CVPR 2022 (Oral) |
|
Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields (jonbarron.info/mipnerf360) Jonathan T. Barron, Ben Mildenhall, Dor Verbin, Pratul P. Srinivasan and Peter Hedman CVPR 2022 (Oral) |
|
Ref-NeRF: Structured View-Dependent Appearance for Neural Radiance Fields (dorverbin.github.io/refnerf) Dor Verbin, Peter Hedman, Ben Mildenhall, Todd Zickler, Jonathan T. Barron and Pratul P. Srinivasan CVPR 2022 (Oral, Best Student Paper Honorable Mention), TPAMI 2024 |
|
HyperNeRF: A Higher-Dimensional Representation for Topologically Varying Neural Radiance Fields (hypernerf.github.io) Keunhong Park, Utkarsh Sinha, Peter Hedman, Jonathan T. Barron, Sofien Bouaziz, Dan B. Goldman, Ricardo Martin-Brualla and Steven M. Seitz SIGGRAPH Asia 2021 |
|
Baking Neural Radiance Fields for Real-Time View Synthesis (nerf.live) Peter Hedman, Pratul P. Srinivasan, Ben Mildenhall, Jonathan T. Barron and Paul Debevec ICCV 2021 (Oral), TPAMI 2024 |
|
Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fields (jonbarron.info/mipnerf) Jonathan T. Barron, Ben Mildenhall, Matthew Tancik, Peter Hedman, Ricardo Martin-Brualla and Pratul P. Srinivasan ICCV 2021 (Oral, Best Paper Honorable Mention) |
|
Immersive Light Field Video with a Layered Mesh Representation (augmentedperception.github.io/deepviewvideo) Michael Broxton, John Flynn, Ryan Overbeck, Daniel Erickson, Peter Hedman, Matthew DuVall, Jason Dourgarian, Jay Busch, Matt Whalen and Paul Debevec SIGGRAPH 2020 |
|
Image-Based Rendering of Cars using Semantic Labels and Approximate Reflection Flow (repo-sam.inria.fr/fungraph/ibr-cars-semantic) Simon Rodriguez, Siddhant Prakash, Peter Hedman and George Drettakis i3D 2020 |
|
Deep Blending for Free-Viewpoint Image-Based Rendering (visual.cs.ucl.ac.uk/pubs/deepblending) Peter Hedman, Julien Philip, True Price, Jan-Michael Frahm, George Drettakis and Gabriel Brostow SIGGRAPH Asia 2018 |
|
Instant 3D Photography (visual.cs.ucl.ac.uk/pubs/instant3d) Peter Hedman and Johannes Kopf SIGGRAPH 2018 |
|
Casual 3D Photography (visual.cs.ucl.ac.uk/pubs/casual3d) Peter Hedman, Suhib Alsisan, Richard Szeliski and Johannes Kopf SIGGRAPH Asia 2017 |
|
Scalable Inside-Out Image-Based Rendering (visual.cs.ucl.ac.uk/pubs/insideout) Peter Hedman, Tobias Ritschel, George Drettakis and Gabriel Brostow SIGGRAPH Asia 2016 |
|
Sequential Monte Carlo Instant Radiosity (visual.cs.ucl.ac.uk/pubs/smcir) Peter Hedman, Tero Karras and Jaakko Lehtinen i3D 2016, TVCG 2017 |
|
Multi-view Reconstruction of Highly Specular Surfaces in Uncontrolled Environments (visual.cs.ucl.ac.uk/pubs/shapefromreflections) Clément Godard*, Peter Hedman*, Wenbin Li and Gabriel J. Brostow 3DV 2015 (Oral) *Joint first authors. |
AWARDS, TALKS AND PUBLICITY
July 2024 | SIGGRAPH Honorable Mention Award for SMERF. |
June 2024 | CVPR XRNeRF workshop talk: "Transcoding for Fast View Synthesis". |
May 2023 | i3D keynote talk: "Scaling NeRF Up and Down". |
June 2022 | CVPR Best Student Paper Honorable Mention Award for Ref-NeRF. |
May 2022 | Our work on NeRF flythroughs for Google Maps presented by Alphabet CEO Sundar Pichai at Google I/O (8:48). |
Oct 2021 | ICCV Best Paper Honorable Mention Award for Mip-NeRF. |
Aug 2020 | SIGGRAPH Immersive Pavilion BEST IN SHOW Award for Immersive Light Field Video with a Layered Mesh Representation. |
Aug 2019 | SIGGRAPH Course (Capture4VR) talk: 3D Photography. |
Jun 2018 | Instant 3D Photography highlighted at the Facebook F8 Conference keynote (1:08:45). |
Apr 2017 | Casual 3D Photography presented by Facebook CEO Mark Zuckerberg at F8 (11:45). |
Nov 2016 | The Finnish Academic Association for Mathematics and Natural Sciences (MAL) prize for the most distinguished Master's thesis. |
Apr 2016 | Rabin Ezra Scholarship for doctoral students in computer graphics, imaging and vision. |
ACADEMIC SERVICE
Area Chair / Papers Committee | CVPR (2024), SIGGRAPH (2023, 2022, session chair), EGSR (2022, 2021, session chair), HPG (2024, 2021) |
Reviewer | SIGGRAPH, SIGGRAPH Asia, CVPR, ECCV, ICCV, TOG, TVCG, 3DV, Computers & Graphics, Eurographics, Pacific Graphics |
Teaching Assistant | Computer Graphics at University College London (2014, 2015, 2017). |
JOBS
May 2024 |
Google DeepMind, Senior Research Scientist, London U.K. Working on Fast View Synthesis. |
May 2022 - May 2024 |
Google, Senior Research Scientist, London U.K. Worked on Fast and Large-Scale Neural Radiance Fields. |
Nov 2020 - May 2022 |
Google, Research Scientist, London U.K. Worked on Fast or Large-Scale Neural Radiance Fields. |
Nov 2019 - Nov 2020 |
Google, Research Scientist, Los Angeles USA. Worked on Immersive Light Field Video with a Layered Mesh Representation (SIGGRAPH 2020). |
Jun 2017 - Aug 2018 |
Pro Unlimited, Contingent worker @ Facebook (Computational photography group), London U.K. Developed Instant 3D Photography (SIGGRAPH 2018). |
Jun 2016 - Jun 2017 |
Facebook, PhD research intern (Computational photography group), Seattle USA. Developed Casual 3D Photography (SIGGRAPH Asia 2017). |
Dec 2013 - May 2014 |
NVIDIA, research intern (NVIDIA research), Helsinki Finland. Developed Sequential Monte Carlo Instant Radiosity (I3D 2016, TVCG 2017). |
Jan 2012 - Nov 2013 |
NVIDIA, Systems software engineer (mobile browser team), Helsinki Finland. Worked with a heavily threaded C++, HTML5 Canvas, WebGL, WebKit, Skia and OpenGL ES2. |
May 2011 - Jan 2012 |
NVIDIA, software engineering intern (Flash3D team), Helsinki Finland. Optimized Flash3D for NVIDIA Tegra. |
EDUCATION
Aug 2014 - Jul 2019 |
PhD in Computer Science (University College London) Viewpoint-Free Photography for Virtual Reality (discovery.ucl.ac.uk/id/eprint/10078087) Supervisors: Gabriel Brostow and Tobias Ritschel. |
Jan 2012 - Jun 2015 |
Master's degree in Computer Science (University of Helsinki) Sequential Monte Carlo Instant Radiosity (helda.helsinki.fi/handle/10138/156669). |
Sep 2009 - Jan 2012 |
Bachelors degree in Computer Science (University of Helsinki) Triangle based and voxel based rendering in real-time graphics (phogzone.com/bscthesis.pdf - in Swedish). |