MSc Thesis Proposal Announcement of Mohammed Farhan Baluch:"Exploring the Effectiveness of Graph Convolutional Networks and Super Pixels for Image Classification"

 

Date: Thursday April 20, 2023

Time:  2:00pm – 3:00pm

Location: Essex Hall, Room 122

 

Reminders: 1. Two-part attendance mandatory (sign-in sheet, QR Code) 2. Arrive 5-10 minutes prior to event starting - LATECOMERS WILL NOT BE ADMITTED. Note that due to demand, if the room has reached capacity, even if you are "early" admission is not guaranteed. 3. Please be respectful of the presenter by NOT knocking on the door for admittance once the door has been closed whether the presentation has begun or not (If the room is at capacity, overflow is not permitted (ie. sitting on floors) as this is a violation of the Fire Safety code). 4. Be respectful of the decision of the advisor/host of the event if you are not given admittance. The School of Computer Science has numerous events occurring soon.

 

 

Super-pixels are a group of pixels that share similar characteristics, reducing the number of data points while preserving visual features, which enables more efficient image analysis. The aim of the research is to investigate whether combining graph convolutional networks (GCNs) with super-pixels can improve the accuracy of image classification, especially in scenarios with limited training data or noisy images. The thesis explores a novel approach that uses super-pixels to reduce the complexity of the image, and then converting it to graphs and applying GCNs to learn the relationships between the super pixels and classify the image. The thesis evaluates the proposed method on multiple datasets and demonstrates its superiority over existing state-of-the-art methods, such as Convolutional Neural Networks (CNN) for image classification when compared for the widely used benchmarks datasets such as MNIST and CIFAR-10. Furthermore, it aims to perform a comparative analysis of various existing super-pixels algorithms, including Simple Linear Iterative Clustering (SLIC) and QuickShift, when used in conjunction with GCNs to achieve superior performance in image classification for the complex datasets such as PASCAL Visual Object Classes (VOC). Overall, this study contributes a new perspective for image classification to the ongoing research in the field of computer vision, and it has the potential to impact real-world applications, such as object recognition, face recognition, and medical image analysis.

 

Keywords: Super-pixels, Graph Convolutional Networks, Image Classification, Computer Vision

 

Internal Reader: Dr. Robin Gras      

External Reader: Dr. Esam Abdel-Raheem    

Advisor: Dr. Luis Rueda