ABSTRACT
A new wave of information is currently underway wherein low-cost 3D scanners and 3D printers combined with popular applications of Computer Graphics and Vision are making 3D object models appeal to an increasing audience. As these 3D object collections are growing, the importance of effective retrieval, search and exploration is increasing. Because it is easier to find something than to create it from scratch and because manual browsing quickly becomes intractable as a collection grows. This talk will strive towards presenting an overview of the research undertaken in Visual Computing Group at Democritus University of Thrace, Greece (VCG – https://vc.ee.duth.gr/ ) addressing challenges on effective 3D Object Classification and Retrieval along a pathway from handcrafted to learned representations.