A Novel Method to Generate Auto-Labeled Datasets for 3D Vehicle Identification Using a New Contrast Model
Auto-labeling is one of the main challenges in 3D vehicle detection. Auto-labeled datasets can be used to identify objects in LiDAR data, which is a challenging task due to the large size of the dataset. In this work, we propose a novel methodology to generate new 3D based auto-labeling datasets wit...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-03-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/13/7/4334 |
_version_ | 1797608337432379392 |
---|---|
author | Guillermo S. Gutierrez-Cabello Edgar Talavera Guillermo Iglesias Miguel Clavijo Felipe Jiménez |
author_facet | Guillermo S. Gutierrez-Cabello Edgar Talavera Guillermo Iglesias Miguel Clavijo Felipe Jiménez |
author_sort | Guillermo S. Gutierrez-Cabello |
collection | DOAJ |
description | Auto-labeling is one of the main challenges in 3D vehicle detection. Auto-labeled datasets can be used to identify objects in LiDAR data, which is a challenging task due to the large size of the dataset. In this work, we propose a novel methodology to generate new 3D based auto-labeling datasets with a different point of view setup than the one used in most recognized datasets (KITTI, WAYMO, etc.). The performance of the methodology has been further demonstrated with the development of our own dataset with the auto-generated labels and tested under boundary conditions on a bridge in a fixed position. The proposed methodology is based on the YOLO model trained with the KITTI dataset. From a camera-LiDAR sensor fusion, it is intended to auto-label new datasets while maintaining the consistency of the ground truth. The performance of the model, with respect to the manually labeled KITTI images, achieves an F-Score of 0.957, 0.927 and 0.740 in the easy, moderate and hard images of the dataset. The main contribution of this work is a novel methodology to auto-label autonomous driving datasets using YOLO as the main labeling system. The proposed methodology is tested under boundary conditions and the results show that this approximation can be easily adapted to a wide variety of problems when labeled datasets are not available. |
first_indexed | 2024-03-11T05:43:03Z |
format | Article |
id | doaj.art-8dc63ddfcc9f42b0a97b0d10be1f73b9 |
institution | Directory Open Access Journal |
issn | 2076-3417 |
language | English |
last_indexed | 2024-03-11T05:43:03Z |
publishDate | 2023-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Applied Sciences |
spelling | doaj.art-8dc63ddfcc9f42b0a97b0d10be1f73b92023-11-17T16:18:51ZengMDPI AGApplied Sciences2076-34172023-03-01137433410.3390/app13074334A Novel Method to Generate Auto-Labeled Datasets for 3D Vehicle Identification Using a New Contrast ModelGuillermo S. Gutierrez-Cabello0Edgar Talavera1Guillermo Iglesias2Miguel Clavijo3Felipe Jiménez4University Institute for Automobile Research (INSIA), Universidad Politécnica de Madrid, 28031 Madrid, SpainDepartamento de Sistemas Informáticos, Escuela Técnica Superior de Ingeniería de Sistemas Informáticos, Universidad Politécnica de Madrid, 28031 Madrid, SpainDepartamento de Sistemas Informáticos, Escuela Técnica Superior de Ingeniería de Sistemas Informáticos, Universidad Politécnica de Madrid, 28031 Madrid, SpainUniversity Institute for Automobile Research (INSIA), Universidad Politécnica de Madrid, 28031 Madrid, SpainUniversity Institute for Automobile Research (INSIA), Universidad Politécnica de Madrid, 28031 Madrid, SpainAuto-labeling is one of the main challenges in 3D vehicle detection. Auto-labeled datasets can be used to identify objects in LiDAR data, which is a challenging task due to the large size of the dataset. In this work, we propose a novel methodology to generate new 3D based auto-labeling datasets with a different point of view setup than the one used in most recognized datasets (KITTI, WAYMO, etc.). The performance of the methodology has been further demonstrated with the development of our own dataset with the auto-generated labels and tested under boundary conditions on a bridge in a fixed position. The proposed methodology is based on the YOLO model trained with the KITTI dataset. From a camera-LiDAR sensor fusion, it is intended to auto-label new datasets while maintaining the consistency of the ground truth. The performance of the model, with respect to the manually labeled KITTI images, achieves an F-Score of 0.957, 0.927 and 0.740 in the easy, moderate and hard images of the dataset. The main contribution of this work is a novel methodology to auto-label autonomous driving datasets using YOLO as the main labeling system. The proposed methodology is tested under boundary conditions and the results show that this approximation can be easily adapted to a wide variety of problems when labeled datasets are not available.https://www.mdpi.com/2076-3417/13/7/4334auto-labeledLiDARpoint of viewdeep learning |
spellingShingle | Guillermo S. Gutierrez-Cabello Edgar Talavera Guillermo Iglesias Miguel Clavijo Felipe Jiménez A Novel Method to Generate Auto-Labeled Datasets for 3D Vehicle Identification Using a New Contrast Model Applied Sciences auto-labeled LiDAR point of view deep learning |
title | A Novel Method to Generate Auto-Labeled Datasets for 3D Vehicle Identification Using a New Contrast Model |
title_full | A Novel Method to Generate Auto-Labeled Datasets for 3D Vehicle Identification Using a New Contrast Model |
title_fullStr | A Novel Method to Generate Auto-Labeled Datasets for 3D Vehicle Identification Using a New Contrast Model |
title_full_unstemmed | A Novel Method to Generate Auto-Labeled Datasets for 3D Vehicle Identification Using a New Contrast Model |
title_short | A Novel Method to Generate Auto-Labeled Datasets for 3D Vehicle Identification Using a New Contrast Model |
title_sort | novel method to generate auto labeled datasets for 3d vehicle identification using a new contrast model |
topic | auto-labeled LiDAR point of view deep learning |
url | https://www.mdpi.com/2076-3417/13/7/4334 |
work_keys_str_mv | AT guillermosgutierrezcabello anovelmethodtogenerateautolabeleddatasetsfor3dvehicleidentificationusinganewcontrastmodel AT edgartalavera anovelmethodtogenerateautolabeleddatasetsfor3dvehicleidentificationusinganewcontrastmodel AT guillermoiglesias anovelmethodtogenerateautolabeleddatasetsfor3dvehicleidentificationusinganewcontrastmodel AT miguelclavijo anovelmethodtogenerateautolabeleddatasetsfor3dvehicleidentificationusinganewcontrastmodel AT felipejimenez anovelmethodtogenerateautolabeleddatasetsfor3dvehicleidentificationusinganewcontrastmodel AT guillermosgutierrezcabello novelmethodtogenerateautolabeleddatasetsfor3dvehicleidentificationusinganewcontrastmodel AT edgartalavera novelmethodtogenerateautolabeleddatasetsfor3dvehicleidentificationusinganewcontrastmodel AT guillermoiglesias novelmethodtogenerateautolabeleddatasetsfor3dvehicleidentificationusinganewcontrastmodel AT miguelclavijo novelmethodtogenerateautolabeleddatasetsfor3dvehicleidentificationusinganewcontrastmodel AT felipejimenez novelmethodtogenerateautolabeleddatasetsfor3dvehicleidentificationusinganewcontrastmodel |