Advancements in Communication and Systems
Unveiling the Power of Graph Neural Network for Intelligent Analysis of Objects
Authors: Prachi Kuldeep Shahane and Shilpa Shinde
Publishing Date: 20-01-2024
ISBN: 978-81-955020-7-3
Abstract
A fundamental and difficult problem in computer vision is object detection. Identifying the visible items in an image may aid in its description and understanding. The extracted data may also be useful for other tasks such as activity detection, content-based picture retrieval, and scene recognition. Every day, billions of people post photographs and videos to the internet as technology and internet access become more widespread. To effectively utilise this enormous amount of data, it is necessary to be able to swiftly and precisely extract information from these pictures. In recent years, substantial advances in object identification and classification have been made possible because of convolutional neural networks (CNN), but it neglects the relationship among objects. In order to enhance object detection performances, this work examined various object detection techniques and used a graph convolutional network (GCN) approach to take advantage of object co-occurrence in an image. Research indicates improved accuracy for object detection and classification.
Keywords
Object detection, Graph Convolutional Networks, Graph Attention Network)
Cite as
Prachi Kuldeep Shahane and Shilpa Shinde, "Unveiling the Power of Graph Neural Network for Intelligent Analysis of Objects", In: Ashish Kumar Tripathi and Vivek Shrivastava (eds), Advancements in Communication and Systems, SCRS, India, 2024, pp. 91-106. https://doi.org/10.56155/978-81-955020-7-3-8