On offline evaluation of vision-based driving models
Codevilla Moraes, Felipe (Centre de Visió per Computador (Bellaterra, Catalunya))
López Peña, Antonio M. (Centre de Visió per Computador (Bellaterra, Catalunya))
Koltun, Vladlen (Intel Corporation)
Dosovitskiy, Alexey (Intel Labs)

Publicación:	Cham, Switzerland: Springer, 2018
Descripción:	17 pàg.
Resumen:	Autonomous driving models should ideally be evaluated by deploying them on a fleet of physical vehicles in the real world. Unfortunately, this approach is not practical for the vast majority of researchers. An attractive alternative is to evaluate models offline, on a pre-collected validation dataset with ground truth annotation. In this paper, we investigate the relation between various online and offline metrics for evaluation of autonomous driving models. We find that offline prediction error is not necessarily correlated with driving quality, and two models with identical prediction error can differ dramatically in their driving performance. We show that the correlation of offline evaluation with driving quality can be significantly improved by selecting an appropriate validation dataset and suitable offline metrics.
Ayudas:	Agencia Estatal de Investigación TIN2017-88709-R Agència de Gestió d'Ajuts Universitaris i de Recerca 2017/FI-B1-00162
Derechos:	Tots els drets reservats.
Lengua:	Anglès
Colección:	Lecture Notes in Computer Science book series (LNCS) ; 11219
Documento:	Capítol de llibre ; recerca ; Versió acceptada per publicar
Materia:	Autonomous driving ; Deep learning
Publicado en:	Computer Vision - ECCV 2018, 2018, p. 246-262, ISBN 978-3-030-01267-0

DOI: 10.1007/978-3-030-01267-0_15

Postprint 17 p, 2.6 MB

El registro aparece en las colecciones:
Libros y colecciones > Capítulos de libros