< Back to Program

Computer Vision and Robot-Vision Systems

Wednesday, 9 August 2023

Lecturers & Schedule

Lazaros Nalpantidis (LN)

Anders Glent Buch (AB)

Dirk Kraft (DK)

08:30 – Welcome and Organizational matters (HP) 

09:00 – 11:30: Perceiving Depth from Images: Stereoscopic Vision and the Stereo Correspondence Problem (LN)

What is Stereo Vision?
Stereo/Epipolar Geometry
Rectified Stereo Case
Depth from Stereo Matches
Correspondence Problem
Dense vs Sparse Correspondence
Local vs Global Correspondence
(Dis)-Similarity Measures

11:30 – Lunch break 

12:30 – 14:30: 6 DoF Object Pose Estimation (AB)

Local 3D descriptors
Transformation estimation using 3D-3D correspondences
Robust pose estimation
Object pose estimation from RGB(D) images
Using deep nets for keypoint and pose prediction

14:30 – 16.00 Exercise

16:00 – 16:30 – Coffee break 

16:30 – 17:30: Computer vision and quality control for robotic assembly (DK)

Challenges in robotic assembly
Computer vision for pick and insert
Computer vision and AI for quality control

17:30 – End of the day