Research
Generally, my research focuses on computational photography and 3D vision. More specifically, I
mainly explore how to utilize various imaging systems, including digital camera, LiDAR, structured
light, and SPAD, etc, to perceive the world and predict its corresponding 3D attributes, such as 3D
geometry, texture, surface material, environment light, and more.
I am particularly interested in how to differentiate imaging systems and embed them within deep
learning frameworks, which allows us to leverage all kinds of optical and physical prior knowledge
inside imaging systems as well as the power of deep learning methods to boost 3D perception
performance.
Thanks to my wonderful collaborators, my research has received several honors.
My work on differentiable rendering was showcased at NVIDIA GTC (Bilibili, Youtube), presented by CEO Jensen Huang (黄仁勋), and has successfully turned into an Omniverse product. The popular YouTube technology channel Two Minute Papers has featured my work multiple times and collected millions of views(DIB-R, GET3D, NVdiffrec). My structured light work won the ICCP 2021 Best Poster Award. My NLOS work was exhibited at Princeton Art of Science Exhibition.
|
News
2024.10 One paper accepted by NeurIPS 2024.
2024.05 Two papers accepted by CVPR 2024 and SIGGRAPH 2024.
2024.05 I have started my AP career at Peking University -:)
|
Hiring
We are actively looking for interns, Masters, and PhDs. Feel free to drop me a line if you are interested in my research or potential collaborations.
For graduate school applicants, I have one Ph.D. opening admitted in 2026.
实验室将招收2025年9月保研, 2026年入学的博士学生,从事LLM+3D Imaging方面的研究,请有意同学提前和我联系。
|
|
TurboSL: Dense, Accurate and Fast 3D by Neural Inverse Structured Light
Parsa Mirdehghan, Maxx Wu,
Wenzheng Chen,
David B. Lindell,
Kiriakos N. Kutulakos
CVPR, 2024
project page
/
video
/
code (coming soon) /
bibtex
TurboSL provides sub-pixel-accurate surfaces and normals at mega-pixel resolution from structured light images, captured at fractions of a second.
|
|
An Expectation-Maximization Algorithm for Training Clean Diffusion Models from Corrupted Observations
Yifei Wang*,
Weimin Bai*,
Wenzheng Chen,
He Sun
(* Equal contribution)
NeurIPS, 2024
arXiv
/
code (coming soon) /
bibtex
EMDiffusion learns a clean diffusion model from corrupted data.
|
|
4D-Rotor Gaussian Splatting: Towards Efficient Novel-View Synthesis for Dynamic Scenes
Yuanxing Duan*,
Fangyin Wei*,
Qiyu Dai,
Yuhang He,
Wenzheng Chen#,
Baoquan Chen#
(* Equal contribution, # joint corresponding authors)
Proc. SIGGRAPH, 2024
arXiv /
code /
bibtex
|
|
Boosting 3D Reconstruction with Differentiable Imaging Systems
Wenzheng Chen
Ph.D. Thesis, 2023
|
|
Flexible Isosurface Extraction for Gradient-Based Mesh Optimization
Tianchang Shen,
Jacob Munkberg,
Jon Hasselgren,
Kangxue Yin,
Zian Wang,
Wenzheng Chen,
Zan Gojcic,
Sanja Fidler,
Nicholas Sharp*,
Jun Gao*
ACM Transactions on Graphics (SIGGRAPH), 2023
project page /
pdf /
supp /
acm /
bibtex
|
|
Neural Fields meet Explicit Geometric Representations for Inverse Rendering of Urban
Scenes
Zian Wang,
Tianchang Shen,
Jun Gao,
Shengyu Huang,
Jacob Munkberg,
Jon Hasselgren,
Zan Gojcic,
Wenzheng Chen,
Sanja Fidler
CVPR, 2023
project page /
arXiv /
codes /
video /
bibtex
Combined with other NVIDIA technology, FEGR is one component of Neural Reconstruction Engine announced in
GTC Sept 2022 Keynote.
|
|
GET3D: A Generative Model of High Quality 3D Textured Shapes Learned from Images
Jun Gao,
Tianchang Shen,
Zian Wang,
Wenzheng Chen,
Kangxue Yin,
Daiqing Li,
Or Litany,
Zan Gojcic,
Sanja Fidler
NeurIPS, 2022
  (Spotlight Presentation)
project page /
arXiv /
codes /
video /
bibtex /
Two Minute Paper
We develop a 3D generative model to generate meshes with textures, bridging the success in the
differentiable surface modeling, differentiable rendering and 2D GANs.
|
|
Neural Light Field Estimation for Street Scenes with Differentiable Virtual Object
Insertion
Zian Wang,
Wenzheng Chen,
David Acuna,
Jan Kautz,
Sanja Fidler
ECCV, 2022
project page /
arXiv /
codes /
video /
bibtex
We propose a hybrid lighting representation to represent spatial-varying lighting for complex
outdoor street scenes.
|
|
Extracting Triangular 3D Models, Materials, and Lighting From Images
Jacob Munkberg,
Jon Hasselgren,
Tianchang Shen,
Jun Gao,
Wenzheng Chen,
Alex Evans,
Thomas Müller,
Sanja Fidler
CVPR, 2022  
(Oral Presentation)
project page /
arXiv /
codes /
video /
bibtex /
Two Minute Paper
Nvdiffrec reconstructs 3D mesh with materials from multi-view images by combining diff surface
modeling with diff renderer. The method supports Nvidia Neural Drivesim
|
|
DIB-R++: Learning to Predict Lighting and Material with a Hybrid Differentiable Renderer
Wenzheng Chen,
Joey Litalien,
Jun Gao,
Zian Wang,
Clement Fuji Tsang,
Sameh Khamis,
Or Litany,
Sanja Fidler
NeurIPS, 2021
project page /
arXiv /
codes /
video /
bibtex
DIB-R++ is a high-performant differentiable renderer which combines rasterization and ray-tracing
together and supports advanced lighitng and material effects. We further embed it in deep learning
and jointly predict geometry, texture, light and material from a single image.
|
|
Image GANs meet Differentiable Rendering for Inverse
Graphics and Interpretable 3D Neural
Rendering
Yuxuan Zhang*, Wenzheng Chen*,
Jun Gao,
Huan Ling, Yinan Zhang,
Antonio Torralba
Sanja Fidler
(* Equal contribution)
ICLR, 2021  
(Oral Presentation)
project page /
arXiv
/ codes /
video /
bibtex
We explore StyleGAN as a multi-view image generator and
train inverse graphics from StyleGAN images. Once trained,
the invere graphics model further helps disentangle and
manipulate StyleGAN latent code from graphics
knowledge. Our work was featured at NVIDIA GTC 2021 and has become an Omniverse product.
|
|
Learned Feature Embeddings for Non-Line-of-Sight
Imaging and Recognition
Wenzheng Chen*,
Fangyin Wei*,
Kyros Kutulakos,
Szymon Rusinkiewicz,
Felix Heide
(* Equal contribution)
SIGGRAPH Asia, 2020  
project page /
paper
/ codes /
bibtex /
Art of Science Exhibition
We propose to learn feature embeddings for
non-line-of-sight imaging and recognition by propagating
features through physical modules.
|
|
Learning Deformable Tetrahedral Meshes for 3D
Reconstruction
Jun Gao,
Wenzheng Chen, Tommy Xiang,
Alec Jacobson,
Morgan Mcguire,
Sanja Fidler
NeurIPS, 2020  
project page /
arXiv
/ codes /
video /
bibtex
We predict deformable tetrahedral meshes from images or
point clouds, which support arbitrary topologies. We also
design a differentiable renderer for tetrahedron, allowing
3D reconstrucion from 2D supervison only.
|
|
Auto-Tuning Structured Light by Optical Stochastic
Gradient Descent
Wenzheng Chen*,
Parsa Mirdehghan*,
Sanja Fidler,
Kyros Kutulakos
(* Equal contribution)
CVPR, 2020  
(ICCP 2021 Best Poster Award)
project page /
paper /
codes /
video /
bibtex
We present optical SGD, a computational imaging technique
that allows an active depth imaging system to
automatically discover optimal illuminations & decoding.
|
|
Learning to Predict 3D Objects with an
Interpolation-based Differentiable Renderer
Wenzheng Chen,
Jun Gao*,
Huan Ling*, Edward J. Smith*,
Jaakko Lehtinen,
Alec Jacobson,
Sanja Fidler
(* Equal contribution)
NeurIPS, 2019  
project page /
arXiv
/
codes
/
bibtex
/
Two Minute Paper
An interpolation-based 3D mesh differentiable renderer
that supports vertex, vertex color, multiple lighting
models, texture mapping and could be easily embedded in
neural networks.
|
|
Steady-state Non-Line-of-Sight Imaging
Wenzheng Chen, Simon Daneau,
Fahim Mannan,
Felix Heide
CVPR, 2019  
(Oral Presentation)
project page /
arXiv
/
codes
/
bibtex
We show hidden objects can be recovereed from conventional images instead of transient images.
|
|
Fast Interactive Object Annotation with
Curve-GCN
Huan Ling*,
Jun Gao*,
Amlan Kar,
Wenzheng Chen,
Sanja Fidler
(* Equal contribution)
CVPR, 2019  
project page /
arXiv
/
codes
/
bibtex
We predict object polygon contours from graph neural
networks, where a novel 2D differentiable rendering loss is
introduced. It renders a polygon countour into a segmentation mask
and back propagates the loss to help optimize the polygon
vertices.
|
|
Optimal Structured Light a la Carte
Parsa Mirdehghan, Wenzheng Chen,
Kyros Kutulakos
CVPR, 2018  
(Spotlight Presentation)
project page /
paper
/
codes
/
bibtex
alacarte designs structured light patterns from a maching
learning persepctive, where patterns are automatically
optimized by minimizing the disparity error under any given imaging condition.
|
|
Synthesizing Training Images for Boosting Human 3D Pose
Estimation
Wenzheng Chen, Huan Wang,
Yangyan Li,
Hao Su,
Zhenhua Wang,
Changhe Tu,
Dani Lischinski,
Daniel Cohen-Or,
Baoquan Chen
3DV, 2016  
(Oral Presentation)
project page /
arXiv
/
codes
/
bibtex
3D pose estimation from model trained with synthetic data
and domain adaptation.
|
|