Yifan Wang
Yifan Wang
Home
News
Talks
Publications
CV
Light
Dark
Automatic
deep learning
Input-level Inductive Biases for 3D Reconstruction
We incorporate inductive biases useful for multiple view geometry into this generalist model without having to touch its architecture, by instead encoding them directly as additional inputs.
Wang Yifan
,
Carl Doersch
,
Relja Arandjelović
, Joao Carreira, Andrew Zisserman
PDF
Cite
Video
Project Page
“Neuralize” geometry processing pipeline
Fueled by the proliferation of consumer-level 3D acquisition devices and the growing accessibility of shape modeling applications for …
Mar 9, 2022 16:30 — 17:30
online
Geometry-Consistent Neural Shape Representation with Implicit Displacement Fields
We extend classic displacement mapping to the neural implicit framework. The resulting novel implicit representation demonstrates superior reconstruction accuracy, parameter efficiency and enable implicit shape editing such as detail transfer.
Wang Yifan
, Lukas Rahmann,
Olga Sorkine-Hornung
PDF
Cite
Code
Video
ICLR Open Review
Video Super-Resolution Using An Artificial Neural Network
According to one implementation, a video processing system includes a computing platform having a hardware processor and a system …
Christopher Schroers
,
Wang Yifan
,
Federico Perazzi
,
Brian McWilliams
,
Alexander Sorkine-Hornung
Iso-Points: Optimizing Neural Implicit Surfaces with Hybrid Representations
Inter-dependent explicit representations for optimizing neural implicit surfaces
Wang Yifan
,
Shihao Wu
,
Cengiz Öztireli
,
Olga Sorkine-Hornung
PDF
Cite
Code
Video
Supplementary
IGL project page
Techniques for performing point-based inverse rendering
Cengiz Öztireli
,
Olga Sorkine-Hornung
,
Shihao Wu
,
Wang Yifan
Techniques for upscaling images generated with undetermined downscaling kernels
Christopher Schroers
,
Wang Yifan
, Victor Cornillère,
Olga Sorkine-Hornung
,
Abdelaziz Djelouah
Detail-Driven 3D Content Creation
A talk summarizing pretty much everything I’ve done in my IGL PhD.
Feb 3, 2021 15:00 — 16:00
online
Follow Toronto Geometry Colloquium
live recording
Pursuing high-resolution 3D Geometry with Deep Learning
Geometric details in 3D shapes is a defining factor in many industries such as AR/VR, VFX and design. However, creating high-fidelity …
Jun 11, 2020 21:00 — 21:30
online
Slides
Neural Cages for Detail-Preserving 3D Deformations
We propose a novel learnable representation for detail-preserving shape deformation extending a traditional cage-based deformation technique. We demonstrate the utility of our method for synthesizing shape variations and deformation transfer.
Wang Yifan
,
Noam Aigerman
,
Vladimir G. Kim
,
Siddhartha Chaudhuri
,
Olga Sorkine-Hornung
PDF
Cite
Code
Supplemental
Talk
IGL Project Page
«
»
Cite
×