Tools
Tools
Pragat, Pravin Pagariya (2024) Lidar – Infused YOLO: A Lidar infused computer vision model to improve the object detection for autonomous vehicle. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (3MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (989kB) | Preview |
Abstract
This research aims at evaluating YOLOv8, YOLOv10, and Lidar-based models in enhancing the precision of 3D objects’ detection in self-driving cars. In automobile detection, YOLOv8 has impressive and fast results; however, it struggles with pedestrian, van, and other objects’ detection, leading to higher false-positive and false-negative rates. YOLOv10 improves the detection accuracy of automobiles and people; however, it is not good at detecting specific objects, including trams and people sitting down. Average accuracies of Frustum PointNets (FPN) are moderate, and better performance is obtained due to the inclusion of fully connected layers, and thus they vary with the level of difficulty. The PointPillers model based on Lidar technology provides high classification accuracy, which makes it 87 percent. 15% and a Loss of 1.72. This effectively separates walkers, bikes, and autos by using 3D bounding boxes in Lidar point cloud. The combined YOLO model using Lidar data with the features of object recognition of YOLO reaches 98% accuracy on the KITTI validation dataset. It is possible to identify item placement and check the correctness of the model with the help of Bird’s Eye View (BEV) photos, which contain information about the possible overlaps and misidentification. The combination of Lidar to YOLO is a new innovation that can be used in real-time 3D object detection for the purpose of self-driving cars. As it stands this technology has the capability for future enhancement and can be applied to many driving scenarios.
| Item Type: | Thesis (Masters) |
|---|---|
| Supervisors: | Name Email Jameel Syed, Muslim UNSPECIFIED |
| Uncontrolled Keywords: | YOLO; LIDAR Infused; Fusion; PointFillers; Bird Eye View; Autonomous Vehicle |
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > TL Motor vehicles. Aeronautics. Astronautics Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence > Computer vision Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence > Computer vision |
| Divisions: | School of Computing > Master of Science in Artificial Intelligence |
| Depositing User: | Ciara O'Brien |
| Date Deposited: | 18 Jun 2025 14:17 |
| Last Modified: | 18 Jun 2025 14:17 |
| URI: | https://norma.ncirl.ie/id/eprint/7922 |
Actions (login required)
 |
View Item |