BEV Object Tracking for LIDAR-based Ground Truth Generation

Autores: David Montero Martín Nerea Aranjuelo Ansa Orti Senderos Yeregui Marcos Nieto Doncel

Fecha: 02.09.2019


Abstract

Building ADAS (Advanced Driver Assistance Systems) or AD (Autonomous Driving) vehicles implies the acquisition of large volumes of data and a costly annotation process to create labeled metadata. Labels are then used for either ground truth composition (for test and validation of algorithms) or to set-up training datasets for machine learning processes. In this paper we present a 3D object tracking mechanism that operates on detections from point cloud sequences. It works in two steps: first an online phase which runs a Branch and Bound algorithm (BBA) to solve the association between detections and tracks, and a second filtering step which adds the required temporal smoothness. Results on KITTI dataset show the produced tracks are accurate and robust against noisy and missing detections, as produced by state-of-the-art deep learning detectors.

BIB_text

@Article {
title = {BEV Object Tracking for LIDAR-based Ground Truth Generation},
pages = {1-5},
keywds = {
Building ADAS (Advanced Driver Assistance Systems) or AD (Autonomous Driving) vehicles implies the acquisition of large volumes of data and a costly annotation process to create labeled metadata. Labels are then used for either ground truth composition (f
}
abstract = {

Building ADAS (Advanced Driver Assistance Systems) or AD (Autonomous Driving) vehicles implies the acquisition of large volumes of data and a costly annotation process to create labeled metadata. Labels are then used for either ground truth composition (for test and validation of algorithms) or to set-up training datasets for machine learning processes. In this paper we present a 3D object tracking mechanism that operates on detections from point cloud sequences. It works in two steps: first an online phase which runs a Branch and Bound algorithm (BBA) to solve the association between detections and tracks, and a second filtering step which adds the required temporal smoothness. Results on KITTI dataset show the produced tracks are accurate and robust against noisy and missing detections, as produced by state-of-the-art deep learning detectors.


}
isbn = {978-9-0827-9703-9},
date = {2019-09-02},
}
Vicomtech

Parque Científico y Tecnológico de Gipuzkoa,
Paseo Mikeletegi 57,
20009 Donostia / San Sebastián (España)

+(34) 943 309 230

Zorrotzaurreko Erribera 2, Deusto,
48014 Bilbao (España)

close overlay

Las cookies de publicidad comportamental son necesarias para cargar el contenido

Aceptar cookies de publicidad comportamental