Accelerating Image Classification using Delaunay Triangulation and Graph Neural Network
PhD Thesis Defense by:
Mustafa Mohammadi Gharasuie
Date: April 29th, 2026
Time: 10AM
Location: Essex Hall 122
Abstract:
Recent advances in image classification have been significantly propelled by the integration of Graph Convolutional Networks (GCNs), offering a novel paradigm for handling complex data structures. This study introduces an innovative framework that employs GCNs in conjunction with Voronoi diagrams to perform image classification, leveraging their exceptional capability to model relational data. Unlike Convolutional neural networks, our approach utilizes a graph-based representation of images, where group of pixels or regions are treated as vertices of a graph, which are then simplified in the form of the corresponding Delaunay triangulations.
Our model yields significant improvement in pre-processing time and classification accuracy on several benchmark datasets, achieves competitive performance compared with previous methods, especially in scenarios that involve complex scenes and fine-grained categories. The experimental results, validated via k-fold cross-validation, underscore the potential of integrating GCNs with Voronoi diagrams in advancing image classification tasks. This research contributes to the field by introducing a novel approach to image classification, while opening new avenues for developing graph-based learning paradigms in other domains of computer vision and non-structured data. In particular, we propose a faster version of GCN, namely normalized Voronoi Graph Convolutional Network (NVGCN).
Keywords: Computer Vision, Vision Transformers, Delaunay Triangulation, Graph Neural Networks
Thesis Committee:
Internal Reader: Prof. Alioune Ngom, Prof. Dan Wu
External Reader: Prof. Mahdi Monfared
Advisor: Prof. Luis Rueda
Chair: Dr. Chitra Rangan
