Seminar on Recent Trends in 3D Robot Vision and Object Recognition

Venue: SEECS Seminar Hall
Time: Friday 24th April, 10:00am - 11:00am

Nowadays object recognition is a fundamental capability for an autonomous robot in order to interact with the physical world. Taking advantage of the new sensing technologies capable of providing RGB-D data, the object recognition capabilities increase dramatically. Object recognition has been well studied; however a problem with object classifiers is that they work properly on the data they are trained but generalize poorly to data from other domains. In the context of this work, a “domain” may refer to data of a certain type, from a certain source, or generated in a certain period in time. Considering this problem in object detectors, a domain adaptation method should be applied. This domain adaptation aims to build a classifier that is robust to the mismatched distributions between different domains. Some domain adaptation approaches have been presented but little work has been done on understanding the effects of domain change on RGB-D data.

Speaker's Profile

Dr. Haider Ali is currently working as a Senior Researcher at the Robotics and Mechatronics Center (RMC) of the German Aerospace Center (DLR). He received his Ph.D. in Computer Vision from Technical University of Vienna, Austria in 2010. His research mainly focuses on the development of robust 3D object recognition and poses estimation methods for real time robotics applications. Over the past 10 years he has proposed and developed a number of innovative methods making a significant contribution in 3D robot vision applications.