RGB-D depth cameras have the potential to become a key component for solving real-world problems. Their low cost and widespread availability have made them a commercial success and their popularity in the research community has dramatically increased. In the meanwhile, the next generation of RGB-D sensors have been developed that are better than current sensors in terms of depth quality, frame-rate, and sensor size. It is therefore a good moment to summarize what has been achieved so far and to discuss interesting possible direction for future work. The proposed short course intends to discuss the basics, underlying principles and cutting-edge results of a comprehensive list of topics in RGB-D perception.

Schedule:

09:15-09:20 Introduction

09:20-10:00 Reconstruction and Interaction

10:00-10:20 Coffee break

10:20-11:10 Scene understanding (slides)

11:10-12:00 RGB-D features and object recognition (slides)

12:00-13:45 Lunch break

13:45-14:30 Pose estimation (slides)

14:30-15:15 Commercial SDKs and APIs (slides)

15:15-15:45 Coffee break

15:45-16:30 Face analysis (slides)

16:30-17:15 Action recognition (slides)