Search Results for author: Adam Lilja

Found 4 papers, 2 papers with code

Exploring Semi-Supervised Learning for Online Mapping

no code implementations14 Oct 2024 Adam Lilja, Erik Wallin, Junsheng Fu, Lars Hammarstrand

The performance gap to using all labels is reduced from 29. 6 to 3. 4 mIoU on Argoverse, and from 12 to 3. 4 mIoU on NuScenes utilising only 10% of the labelled data.

Autonomous Driving

Are NeRFs ready for autonomous driving? Towards closing the real-to-simulation gap

no code implementations24 Mar 2024 Carl Lindström, Georg Hess, Adam Lilja, Maryam Fatemi, Lars Hammarstrand, Christoffer Petersson, Lennart Svensson

Specifically, we evaluate object detectors and an online mapping model on real and simulated data, and study the effects of different fine-tuning strategies. Our results show notable improvements in model robustness to simulated data, even improving real-world performance in some cases.

Autonomous Driving Data Augmentation +3

Localization Is All You Evaluate: Data Leakage in Online Mapping Datasets and How to Fix It

1 code implementation CVPR 2024 Adam Lilja, Junsheng Fu, Erik Stenborg, Lars Hammarstrand

Specifically, over $80$% of nuScenes and $40$% of Argoverse 2 validation and test samples are less than $5$ m from a training sample.

Zenseact Open Dataset: A large-scale and diverse multimodal dataset for autonomous driving

1 code implementation ICCV 2023 Mina Alibeigi, William Ljungbergh, Adam Tonderski, Georg Hess, Adam Lilja, Carl Lindstrom, Daria Motorniuk, Junsheng Fu, Jenny Widahl, Christoffer Petersson

The dataset is composed of Frames, Sequences, and Drives, designed to encompass both data diversity and support for spatio-temporal learning, sensor fusion, localization, and mapping.

Autonomous Driving Diversity +4

Cannot find the paper you are looking for? You can Submit a new open access paper.