Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ji, Yishen, Li, Zhiqi, Lu, Tong
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2412.16938
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916538495270912
author Ji, Yishen
Li, Zhiqi
Lu, Tong
author_facet Ji, Yishen
Li, Zhiqi
Lu, Tong
contents Track Mapless demands models to process multi-view images and Standard-Definition (SD) maps, outputting lane and traffic element perceptions along with their topological relationships. We propose a novel architecture that integrates SD map priors to improve lane line and area detection performance. Inspired by TopoMLP, our model employs a two-stage structure: perception and reasoning. The downstream topology head uses the output from the upstream detection head, meaning accuracy improvements in detection significantly boost downstream performance.
format Preprint
id arxiv_https___arxiv_org_abs_2412_16938
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ImagineMap: Enhanced HD Map Construction with SD Maps
Ji, Yishen
Li, Zhiqi
Lu, Tong
Computer Vision and Pattern Recognition
Track Mapless demands models to process multi-view images and Standard-Definition (SD) maps, outputting lane and traffic element perceptions along with their topological relationships. We propose a novel architecture that integrates SD map priors to improve lane line and area detection performance. Inspired by TopoMLP, our model employs a two-stage structure: perception and reasoning. The downstream topology head uses the output from the upstream detection head, meaning accuracy improvements in detection significantly boost downstream performance.
title ImagineMap: Enhanced HD Map Construction with SD Maps
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.16938