all AI news
EAN-MapNet: Efficient Vectorized HD Map Construction with Anchor Neighborhoods
Feb. 29, 2024, 5:45 a.m. | Huiyuan Xiong, Jun Shen, Taohong Zhu, Yuelong Pan
cs.CV updates on arXiv.org arxiv.org
Abstract: High-definition (HD) map is crucial for autonomous driving systems. Most existing works design map elements detection heads based on the DETR decoder. However, the initial queries lack integration with the physical location feature of map elements, and vanilla self-attention entails high computational complexity. Therefore, we propose EAN-MapNet for Efficiently constructing HD map using Anchor Neighborhoods. Firstly, we design query units based on the physical location feature of anchor neighborhoods. Non-neighborhood central anchors effectively assist the …
abstract anchor arxiv attention autonomous autonomous driving autonomous driving systems complexity computational construction cs.cv decoder definition design detection detr driving feature integration location map queries self-attention systems type
More from arxiv.org / cs.CV updates on arXiv.org
Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMs
2 days, 12 hours ago |
arxiv.org
Jobs in AI, ML, Big Data
Data Architect
@ University of Texas at Austin | Austin, TX
Data ETL Engineer
@ University of Texas at Austin | Austin, TX
Lead GNSS Data Scientist
@ Lurra Systems | Melbourne
Senior Machine Learning Engineer (MLOps)
@ Promaton | Remote, Europe
Research Scientist, Demography and Survey Science, University Grad
@ Meta | Menlo Park, CA | New York City
Computer Vision Engineer, XR
@ Meta | Burlingame, CA