Title: A study of object recognition and tracking techniques for augmented reality applications

Authors: Chandrasekar Ravi; Manoja Talagadadeevi

Addresses: Department of Computer Science and Engineering, National Institute of Technology Puducherry, India ' Department of Computer Science and Engineering, National Institute of Technology Puducherry, India

Abstract: Augmented reality (AR) overlays 3D graphics or 2D computer-generated information over the user's field of view of the real world's surroundings, thereby improving the user's capability to perceive the real world. To recognise an object from the real world and estimate its location, various methods are carried out, such as object recognition and tracking. This paper provides a comprehensive study of various object recognition and tracking algorithms for AR applications. Object recognition algorithms capture and detect objects from a real environment, whereas object tracking algorithms track the movement of the object. This paper discusses how AR can be implemented using different hardware devices, predefined software development toolkits (SDKs), and various computer vision algorithms that can be developed for object recognition and tracking with their corresponding limitations. It concludes by presenting a case study on an augmented reality assisted assembly for the manufacturing sector.

Keywords: augmented reality; computer vision; object recognition; object tracking; software development toolkits; SDKs.

DOI: 10.1504/IJCVR.2024.141870

International Journal of Computational Vision and Robotics, 2024 Vol.14 No.6, pp.589 - 614

Received: 16 Jul 2022
Accepted: 15 Mar 2023
Published online: 02 Oct 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article