Organizers
Sean Fanello, Julien Valentin, Jonathan Taylor, Christoph Rhemann, Adarsh Kowdle, Jürgen Sturm,
Christine Kaeser-Chen, Pavel Pidlypenskyi, Rohit Pandey, Andrea Tagliasacchi, Sameh Khamis,
David Kim, Mingsong Dou, Kaiwen Guo, Danhang Tang, Sofien Bouaziz, Shahram Izadi
Description
The emergence of virtual and augmented reality has increased the demand of robust systems for 3D capture, reconstruction and understanding. Designing such systems involves developing high quality sensors and efficient algorithms that can leverage new and existing technologies. To this end, we design depth sensors with two properties that drastically simplify the problem of fusing incomplete sensor data. First, an UltraFast depth stream that significantly reduces the frame to frame motion. Second, by allowing multiple sensors to be easily combined (without interference) we can eliminate occlusions. As a result, we have developed a wide range of efficient algorithms for reconstruction, tracking and understanding that are designed to work with this technology. In this tutorial we will walk the audience through the ins and outs of building such a stack from the ground up. From building such a sensor to applications for mixed reality with particular emphasis on high speed 3D capture systems.
When and Where
September 8th, Technische Universität München (TU), Arcisstraße 21 80333 Munich
Program
Time |
Title |
Speaker |
|---|---|---|
9:00 - 9:15 |
Introduction |
Shahram Izadi |
Morning Session: Depth Sensors, 3D Capture & Camera Tracking |
||
9:15 - 9:45 |
Depth Sensors and Algorithms: What, When, Where |
Sean Fanello |
9:45 - 10:15 |
Triangulation Methods: Basics and Challenges |
Christoph Rhemann |
10:15 - 10:30 |
Low Compute and Fully Parallel Computer Vision with HashMatch |
Julien Valentin |
10:30 - 11:00 |
Coffee Break |
|
11:00 - 11:15 |
Depth Completion of a Single RGB-D Image |
Yinda Zhang Princeton University / Google |
11:15 - 11:35 |
Depth Estimation in the Age of Deep Learning |
Sameh Khamis |
11:35 - 11:50 |
Learning a Multi-View Stereo Machine |
Christian Haene |
11:50 - 12:30 |
Localization and Mapping - ARCore |
Konstantine Tsotsos |
12:30 - 13:30 |
Lunch Break |
|
Afternoon Session: World and Human Reconstruction and Understanding |
||
13:30 - 13:45 |
3D Scene Reconstruction |
Thomas Funkhouser Princeton University / Google |
13:45 - 14:00 |
3D Scene Understanding |
Martin Bokeloh |
14:00 - 14:15 |
Coffee Break |
|
14:15 - 14:45 |
Non-Linear Optimization Methods |
Andrea Tagliasacchi |
14:45 - 15:00 |
Parametric Tracking of Hands |
Anastasia Tkach |
15:00 - 15:15 |
Parametric Tracking of Faces |
Sofien Bouaziz |
15:15 - 15:30 |
Coffee Break |
|
15:30 - 16:00 |
Non Rigid Reconstruction of Humans |
Kaiwen Guo |
16:00 - 16:15 |
Real-time Compression and Streaming of 4D Performances |
Danhang Tang |
16:15 - 16:30 |
Neural rendering for Performance Capture |
Ricardo Martin Brualla |
Please contact Shahram Izadi or Sean Fanello if you have any questions or to request the slides.