Title: A survey on autonomous vehicle: recent advancements in 2D/3D lane and object detection algorithms

Authors: R. Rajesh; P.V. Manivannan

Addresses: Department of Mechanical Engineering, Indian Institute of Technology Madras, Chennai-600036, India ' Department of Mechanical Engineering, Indian Institute of Technology Madras, Chennai-600036, India

Abstract: This paper provides a comprehensive analysis of advanced algorithms for autonomous vehicles (AVs), with a focus on lane detection, object detection, and evaluation datasets. In the first part, both traditional and deep learning-based 2D/3D lane detection approaches are discussed in detail with their methodologies and applicability. In addition, the lane detection approaches are categorised into traditional and deep learning-based approaches for 2D/3D lane detection, highlighting their characteristics and performance. Subsequently, in the second part, the study then shifts to 2D/3D traditional and deep learning-based object detection, including birds-eye-view detection and transformer network-based approaches, offering a critical analysis of these techniques and their potential impact on autonomous driving technology. Finally, an overview of state-of-the-art datasets for evaluating lane and object detection algorithms in autonomous driving research is presented. Overall, this survey consolidates the current state of research in lane and object detection algorithms for autonomous vehicles, providing insights and directions for future research in this field.

Keywords: autonomous driving; bird's eye view detector; convolutional neural network; CNN; deep learning; Hough transform; lane detection; object detection; sensor fusion; transformer networks.

DOI: 10.1504/IJVP.2026.152842

International Journal of Vehicle Performance, 2026 Vol.12 No.1, pp.98 - 146

Received: 31 Mar 2025
Accepted: 03 Oct 2025
Published online: 13 Apr 2026 *

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