Generative Models for Unstructured 3D Shapes

Lead Research Organisation: University College London
Department Name: Computer Science

Abstract

Neural networks have specially been designed for unstructured data (e.g., point clouds, meshes). Such networks have mainly been used for analysing unstructured 3D data, either for classification and/or part segmentation. This thesis will develop new learning frameworks to facilitate synthesis of novel shapes by considering the inverse modelling problem.

Publications

10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/S021566/1 01/04/2019 30/09/2027
2253986 Studentship EP/S021566/1 23/09/2019 22/09/2023 Luca Morreale