Computer vision-based reconstructive plastic surgery

High energy traumatic impact of the craniofacial skeleton is an inevitable consequence of todays fast paced society. The work in the thesis leverages recent advances in computer vision, computer visualization and computer aided design/manufacturing to reduce the fractures and reconstruct the craniofacial skeleton in silico. A graphical user interface is designed to help the surgeon to perform the surgery in a virtual environment. Surface matching algorithms such as the Iterative Closest Point (ICP) and the Data-Aligned Rigidity-Constrained Exhaustive Search (DARCES) are applied separately to the problem of craniofacial reconstruction by matching the fragment surfaces. A synergetic combination of the DARCES and ICP algorithms where the output of the DARCES algorithm is fed as input to the ICP algorithm results in an improved performance in terms of both reconstruction accuracy and execution time.