U-BEV: Height-aware Bird's-Eye-View Segmentation and Neural Map-based Relocalization

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Camiletto, Andrea Boscolo, Bochicchio, Alfredo, Liniger, Alexander, Dai, Dengxin, Gawel, Abel
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908288922157056
author Camiletto, Andrea Boscolo
Bochicchio, Alfredo
Liniger, Alexander
Dai, Dengxin
Gawel, Abel
author_facet Camiletto, Andrea Boscolo
Bochicchio, Alfredo
Liniger, Alexander
Dai, Dengxin
Gawel, Abel
contents Efficient relocalization is essential for intelligent vehicles when GPS reception is insufficient or sensor-based localization fails. Recent advances in Bird's-Eye-View (BEV) segmentation allow for accurate estimation of local scene appearance and in turn, can benefit the relocalization of the vehicle. However, one downside of BEV methods is the heavy computation required to leverage the geometric constraints. This paper presents U-BEV, a U-Net inspired architecture that extends the current state-of-the-art by allowing the BEV to reason about the scene on multiple height layers before flattening the BEV features. We show that this extension boosts the performance of the U-BEV by up to 4.11 IoU. Additionally, we combine the encoded neural BEV with a differentiable template matcher to perform relocalization on neural SD-map data. The model is fully end-to-end trainable and outperforms transformer-based BEV methods of similar computational complexity by 1.7 to 2.8 mIoU and BEV-based relocalization by over 26% Recall Accuracy on the nuScenes dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2310_13766
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle U-BEV: Height-aware Bird's-Eye-View Segmentation and Neural Map-based Relocalization
Camiletto, Andrea Boscolo
Bochicchio, Alfredo
Liniger, Alexander
Dai, Dengxin
Gawel, Abel
Computer Vision and Pattern Recognition
Efficient relocalization is essential for intelligent vehicles when GPS reception is insufficient or sensor-based localization fails. Recent advances in Bird's-Eye-View (BEV) segmentation allow for accurate estimation of local scene appearance and in turn, can benefit the relocalization of the vehicle. However, one downside of BEV methods is the heavy computation required to leverage the geometric constraints. This paper presents U-BEV, a U-Net inspired architecture that extends the current state-of-the-art by allowing the BEV to reason about the scene on multiple height layers before flattening the BEV features. We show that this extension boosts the performance of the U-BEV by up to 4.11 IoU. Additionally, we combine the encoded neural BEV with a differentiable template matcher to perform relocalization on neural SD-map data. The model is fully end-to-end trainable and outperforms transformer-based BEV methods of similar computational complexity by 1.7 to 2.8 mIoU and BEV-based relocalization by over 26% Recall Accuracy on the nuScenes dataset.
title U-BEV: Height-aware Bird's-Eye-View Segmentation and Neural Map-based Relocalization
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2310.13766