Benchmark

  • Click here to find the result of the preliminary round of this competition.
  • Click here to download the result of the final round of this competition.
  • Configuration of the benchmark PC:
    • CPU : i7-12700F 4.90GHz
    • Memory : 32G
    • GPU : NVIDIA RTX 3060-12G
    • Hard Disk : NVME SSD 1TB
Rank Participants Group Name Affiliation System Name APE RPE ARE RRE Badness Initialization Quality Robustness Relocalization Time Benchmark Score Average FrameRate Speed Penalty Final Score (%) System Description
1 冯杰琳,董宇祥,邹征昊,赵李群,赵春晖,胡劲文,吕洋 李群不会漂移 西北工业大学 LIFT-VINS 0.2051 0.1372 0.0998 0.0778 0.0294 0.8697 0.4941 0.9764 0.4637 29.9918 0.9997 46.36 Doc Slide
2 张维智,颜长建,盛兴东 Lenovo Research 联想上海研究院 AR-SLAM 0.1222 0.1753 0.0569 0.0807 0.0025 0.8357 0.3926 0.8619 0.4000 32.4606 1.0000 40.00 Doc Slide
3 徐泽文,吕泽仁,张怡迪,杨钊龙,卫浩 薇爱欧 中国科学院自动化研究所 RVG-VIO-Opti 0.2360 0.1729 0.1081 0.0800 0.0776 0.8591 0.6061 0.3141 0.3879 66.9793 1.0000 38.79 Doc Slide
4 张雪涛,林舒月,姚贺凯,徐啸天,孙刚 DUTSLAM 大连理工大学 DUT-VISLAM 0.0972 0.3373 0.0512 0.0735 0.0034 0.5957 0.4233 0.9134 0.3821 30.0195 1.0000 38.21 Doc Slide
5 吴昱臻,白宇,张廉,王岭雪 全向感知 北京理工大学 VINS-wo-ROS 0.0548 0.1633 0.0359 0.0611 0.0090 0.3719 0.1788 0.9871 0.2903 28.4567 0.9486 27.54 Doc Slide

Competition Chairs


Acknowledgement


We thank Ziyang Zhang for his great help in building the website and evaluating the participating SLAM systems. Similarly, our thanks go to Jin Yuan for his support in recording the dataset and evaluating the participating SLAM systems.