Paper Key : IRJ************421
Author: Mantu Naresh Sharma
Date Published: 22 Oct 2023
Abstract
In this paper, we evaluated the performance of computer vision and machine learning has revolutionized the field of image and video processing, bringing in a new era of unparalleled creativity and insightful applications, that are evaluated in this research. Despite the many advances in this domain, object identification and segmentation have become essential technologies with a wide range of applications, from improving surveillance to enhancing autonomous car safety to boosting augmented reality experiences. YOLO-NAS (You Only Look Once Neural Architecture Search) and the Segment Anything Model (SAM), two leading-edge models are utilized in this research to delve into the evolving arena of image and video segmentation. In real-time detection of objects and object delineation, YOLO-NAS and SAM models establish their mutual benefit and promise as an integrated approach to image and video segmentation in an evolving digital arena.