IROS 2019终身机器视觉挑战赛——终身物体识别报告
计算机视觉与模式识别
2020-05-01 v1 机器学习
机器人学
图像与视频处理
机器学习
摘要
本报告总结了IROS 2019终身机器视觉竞赛（终身物体识别挑战赛），包含来自前名决赛队伍（超过支队伍）的方法与结果。竞赛数据集(L)ifel(O)ng (R)obotic V(IS)ion (OpenLORIS)-Object Recognition (OpenLORIS-object)旨在推动机器人视觉领域终身/持续学习的研究与应用，包含家庭、办公室、校园和商场场景中的日常物体。该数据集明确量化了光照、物体遮挡、物体尺寸、相机-物体距离/角度以及杂波信息的变异情况。规则被设计用于量化机器视觉系统在面对竞赛中动态环境出现的物体时的学习能力。各参赛队伍报告、数据集信息、规则及发布的源代码可在项目主页找到："https://lifelong-robotic-vision.github.io/competition/"。
引用
@article{arxiv.2004.14774,
title = {IROS 2019 Lifelong Robotic Vision Challenge -- Lifelong Object Recognition Report},
author = {Qi She and Fan Feng and Qi Liu and Rosa H. M. Chan and Xinyue Hao and Chuanlin Lan and Qihan Yang and Vincenzo Lomonaco and German I. Parisi and Heechul Bae and Eoin Brophy and Baoquan Chen and Gabriele Graffieti and Vidit Goel and Hyonyoung Han and Sathursan Kanagarajah and Somesh Kumar and Siew-Kei Lam and Tin Lun Lam and Liang Ma and Davide Maltoni and Lorenzo Pellegrini and Duvindu Piyasena and Shiliang Pu and Debdoot Sheet and Soonyong Song and Youngsung Son and Zhengwei Wang and Tomas E. Ward and Jianwen Wu and Meiqing Wu and Di Xie and Yangsheng Xu and Lin Yang and Qiaoyong Zhong and Liguang Zhou},
journal= {arXiv preprint arXiv:2004.14774},
year = {2020}
}
备注
9 pages, 11 figures, 3 tables, accepted into IEEE Robotics and Automation Magazine. arXiv admin note: text overlap with arXiv:1911.06487