3D Object Detection For Intelligent Vehicles

3D Object Detection For Intelligent Vehicles

van der Sluis, Joram (TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Intelligent Vehicles; TU Delft Cognitive Robotics)

Gavrila, D. (mentor)
Pool, E.A.I. (mentor)
Pan, W. (graduation committee)
van Gemert, J.C. (graduation committee)

Delft University of Technology

Mechanical Engineering | Vehicle Engineering

2020-10-26

This master thesis presents an experimental study on 3D person localization (i.e., pedestrians, cyclists)in traffic scenes, using monocular vision and Light Detection And Ranging (LiDAR) data. The performance of two top-ranking methods is analyzed on the 3D object detection KITTI dataset. In this evaluation, the effect of the Intersection over Union (IoU) threshold on the performance in terms of 3D bounding box location, size, and orientation is analysed.
Since the KITTI 3D object detection dataset contains relatively few 3D person instances, the analysis will is to the EuroCity Persons 2.5D (ECP2.5D) datasets (both day and night), which is one order of magnitude larger. Using both datasets, additional experiments are performed to evaluate the influence of distance, the number of LiDAR points, occlusion, and intensity on the performance. Domain transfer experiments between the KITTI and ECP2.5D datasets are performed, to examine how these datasets generalize with respect to each other. Furthermore, Part-A2 net is used to evaluate the detection score which is given to the ground truth pedestrians. The relationship between the detection score and the distance, the number of LiDAR points, and occlusion is analyzed. Some objects are not detected although their ground truth detection score is high. This creates the potential to detect these pedestrians. Lastly, this thesis presents a method that uses the detections from the previous frame to increase the performance in the subsequent frame by adding the previous detections to the 3D proposals coming from the Region Proposal Network (RPN).

3D object detection
Pedestrian detection
Vulnerable Road User
LiDAR
Camera
Intelligent Vehicles
Robotics
Point Cloud
Temporal fusion
3D Person Localization
Intersection over Union
KITTI dataset
EuroCity Persons 2.5D dataset
LiDAR Points
Occlusion
Intensity
Distance
Experimental study

http://resolver.tudelft.nl/uuid:55819905-53fb-41bd-839f-6e840216035a

During this master thesis, two articles were written. The first article will be presented at the 2020 IEEE Intelligent Vehicles Symposium (IV2020) titled “An Experimental Study on 3D Person Localization in Traffic Scenes” by Joram R. van der Sluis, Ewoud A.I. Pool, and Dariu M. Gavrila. The second article will be submitted to IEEE Transactions on Intelligent Vehicles (T-IV) titled “An Experimental Study on 3D Person Localization in Traffic Scenes” by Joram R. van der Sluis, Ewoud A.I. Pool, and Dariu M. Gavrila. This second article is an extension of the first article. Since this article is not submitted yet, it could be that the appended version will contain some differences regarding the final submitted version.

Student theses

master thesis

© 2020 Joram van der Sluis