| Title |
DINO pre-training for vision-based end-to-end autonomous driving / |
| Authors |
Juneja, Shubham ; Daniušis, Povilas ; Marcinkevičius, Virginijus |
| DOI |
10.22364/bjmc.2024.12.4.02 |
| Full Text |
|
| Is Part of |
Baltic journal of modern computing.. Riga : University of Latvia. 2024, vol. 12, no. 4, p. 374-386.. ISSN 2255-8942. eISSN 2255-8950 |
| Keywords [eng] |
autonomous driving ; DINO ; self-supervised pre-training |
| Abstract [eng] |
In this article, we focus on the pre-training of visual autonomous driving agents in the context of imitation learning. Current methods often rely on a classification-based pre-training, which we hypothesise to be holding back from extending capabilities of implicit image understanding. We propose pre-training the visual encoder of a driving agent using the self-distillation with no labels (DINO) method, which relies on a self-supervised learning paradigm.Our experiments in CARLA environment in accordance with the Leaderboard benchmark reveal that the proposed pre-training is more efficient than classification-based pre-training, and is on par with the recently proposed pre-training based on visual place recognition (VPRPre). |
| Published |
Riga : University of Latvia |
| Type |
Journal article |
| Language |
English |
| Publication date |
2024 |
| CC license |
|
