A New Benchmark Model for Monocular 3D Lane Detection in Autonomous Driving using Transformative Architecture

Handoko, Mohammed Fachry Dwi (2025) A New Benchmark Model for Monocular 3D Lane Detection in Autonomous Driving using Transformative Architecture. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5025201159-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Available under License Creative Commons Attribution Non-commercial Share Alike. Download (3MB) | Request a copy

Abstract

Three-dimensional lane detection and modeling is a technique that uses deep learning algorithms to detect and visualize lanes in a 3D space using a transformative approach. Various published models have shown it to be effective in detecting and modeling lanes in 3D space, with the latest benchmark having been achieved by the LATR model (F1 = 61.9). This technology has numerous potential applications in autonomous driving, robotics, and other fields, being designed to perceive lanes that provide spatial guidance for machines. This project seeks to forward an improved model with the capability to outperform previous lane detection models in terms of F-score, accuracy, and or epoch. With more refined techniques for convolutional neural networks (CNNs) in extracting features from an input image, followed by a modified transformer to predict the 3D coordinates of the lane markers from a flat ground plane, the new and highly derived model managed to almost match the limitations set by LATR itself and achieve a maximum F-score of 0.5561.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	3D, CNN, neural network, autonomous, transformer, lane, deep learning
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > Q Science (General) > Q337.5 Pattern recognition systems T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.83 Dynamic programming
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	MOHAMMED FACHRY DWI HANDOKO
Date Deposited:	03 Feb 2025 02:08
Last Modified:	03 Feb 2025 02:08
URI:	http://repository.its.ac.id/id/eprint/117808

Actions (login required)

View Item