VMA: Divide-and-Conquer Vectorized Map Annotation System for Large-Scale Driving Scene

ArXiv Preprint (arXiv 2304.09807)

News

• Aug. 30th, 2023: We release an initial version of VMA.
• Aug. 9th, 2023: Code will be released in around 3 weeks.

demo.mp4

TL;DR VMA is a general map auto annotation framework based on MapTR, with high flexibility in terms of spatial scale and element type.

Getting Started

• Installation
• Prepare Dataset
• Inference on SD data (we only provide some samples of SD data for inference, since SD data is owned by Horizon)
• Train and Eval on NYC data

Auto Annotation Results

Remote sensing: Urban scene: Highway scene:

Citation

If you find VMA is useful in your research or applications, please consider giving us a star 🌟 and citing it by the following BibTeX entry.

@inproceedings{VMA,
  title={VMA: Divide-and-Conquer Vectorized Map Annotation System for Large-Scale Driving Scene},
  author={Chen, Shaoyu and Zhang, Yunchi and Liao, Bencheng, Xie, Jiafeng and Cheng, Tianheng and Sui, Wei and Zhang, Qian and Liu, Wenyu and Huang, Chang and Wang, Xinggang},
  booktitle={arXiv preprint arXiv:2304.09807},
  year={2023}
}