HKUCS Computer Vision Lab Virtual Workshop Series
When 3D reconstruction meets deep learning
Host: Dr. Kenneth K.Y. Wong, University of Hong Kong
Date: 07 Jan 2021
Time: 10:30am - 1:50pm
(GMT+8, HK Time)
Webinar via Zoom
introduction
3D reconstruction refers to the process of recovering 3D shape of objects from sensor data. It has always been one of the key topics in computer vision, and has broad applications in robotics, visualization, entertainment, healthcare, etc. Over the past 5 decades, many sophisticated methods have been proposed based on physical and geometric constraints, and the reconstruction problem is once considered largely solved. With the recent success of deep learning in many computer vision tasks, researchers are interested in revisiting the 3D reconstruction problem and deriving novel learning based methods to achieve new standards in reconstruction quality and robustness. This workshop invites internationally renowned experts to share their findings and insights in deep learning based 3D reconstruction with the research community.
Invited Speakers
Prof. Noah Snavely
Cornell University, USA
The Plenoptic Camera
Prof. Hao Li
Pinscreen and UC Berkeley, USA
Making Avatars and Volumetric Teleportation Accessible using 3D Deep Learning
Prof. Tat-Jun Chin
The University of Adelaide, Australia
End-to-End Learnable Geometric Vision
Prof. Kyoung Mu Lee
Seoul National University, Korea
Learning to Reconstruct Whole Expressive 3D Human Pose and Shape from Single Image
Prof. Andreas Geiger
University of Tübingen, Germany
Neural Implicit Representations for 3D Vision
Rundown
| HK Time (GMT+8) | Title of Talk | Speaker |
|---|---|---|
| 10:30am – 10:35am | Opening remarks by Dean of Engineering | Prof. Christopher Chao The University of Hong Kong, HK |
| 10:35am – 11:10am | Talk 1 : The Plenoptic Camera | Prof. Noah Snavely Cornell University, USA |
| 11:10am – 11:45am | Talk 2 : Making Avatars and Volumetric Teleportation Accessible using 3D Deep Learning | Prof. Hao Li Pinscreen and UC Berkeley, USA |
| 11:45am – 11:55am | Break | |
| 11:55am – 12:30pm | Talk 3 : End-to-End Learnable Geometric Vision | Prof. Tat-Jun Chin The University of Adelaide, Australia |
| 12:30pm – 1:05pm | Talk 4 : Learning to Reconstruct Whole Expressive 3D Human Pose and Shape from Single Image | Prof. Kyoung Mu Lee Seoul National University, Korea |
| 1:05pm – 1:15pm | Break | |
| 1:15pm – 1:50pm | Talk 5 : Neural Implicit Representations for 3D Vision | Prof. Andreas Geiger University of Tübingen, Germany |
