Towards Fast, Accurate, Smooth, and General Shape Correspondences

PhD Qualifying Examination


Title: "Towards Fast, Accurate, Smooth, and General Shape Correspondences"

by

Mr. Barakeel FANSEU KAMHOUA


Abstract:

Shape correspondence, or shape matching, is a fundamental task in computer 
graphics with applications in shape registration, interpolation, 
deformation, motion and style transfer, and animation. Achieving fast, 
accurate, smooth, and general shape matching remains a significant challenge 
due to the conflicting requirements of unique and deformation-invariant 
descriptors for accuracy, and smooth and continuous descriptors for 
neighborhood coherence. This survey reviews the challenges in shape 
correspondence, existing approaches, and their trade-offs in terms of 
accuracy, smoothness, generalization, and computational efficiency. We also 
explore a promising approach that leverages k-hop neighborhood descriptors 
combined with local map distortion (LMD) to balance these factors.


Date:                   Tuesday, 27 May 2025

Time:                   2:00pm - 4:00pm

Venue:                  Room 2130A
                        Lift 19

Committee Members:      Prof. Huamin Qu (Supervisor)
                        Prof. Pedro Sander (Chairperson)
                        Dr. Anyi Rao (AMC)