计算机应用 ›› 2021, Vol. 41 ›› Issue (4): 1165-1171.DOI: 10.11772/j.issn.1001-9081.2020071039

所属专题: 多媒体计算与计算机仿真

• 多媒体计算与计算机仿真 • 上一篇    下一篇

基于坐标逆映射的增强型车辆三维全景影像

谭兆一1,2, 陈白帆1   

  1. 1. 中南大学 自动化学院, 长沙 410083;
    2. 伦敦大学学院 计算机科学系, 伦敦 WC1E 6BT, 英国
  • 收稿日期:2020-07-17 修回日期:2020-11-11 出版日期:2021-04-10 发布日期:2021-01-19
  • 通讯作者: 陈白帆
  • 作者简介:谭兆一(1998—),男,陕西宝鸡人,硕士研究生,主要研究方向:图像处理、机器学习、计算机视觉;陈白帆(1979—),女,湖南常德人,副教授,博士,主要研究方向:无人车、环境感知。
  • 基金资助:
    国家重点研发计划项目(2018YFB1201602);湖南省自然科学基金青年基金资助项目(2018JJ3689)。

Enhanced vehicle 3D surround view based on coordinate inverse mapping

TAN Zhaoyi1,2, CHEN Baifan1   

  1. 1. School of Automation, Central South University, Changsha Hunan 410083, China;
    2. Department of Computer Science, University College London, London WC1E 6BT, UK
  • Received:2020-07-17 Revised:2020-11-11 Online:2021-04-10 Published:2021-01-19
  • Supported by:
    This work is partially supported by the National Key Research and Development Program of China (2018YFB1201602), the Youth Program of Natural Science Foundation of Hunan Province (2018JJ3689).

摘要: 当前最先进的车辆三维全景影像虽然可以较好地对车身周边环境进行三维立体的拟真显示,但仍然会对车身近处的三维物体造成显示畸变,极大地影响显示效果、降低实用性。针对该问题,提出一种增强型车辆三维全景影像的合成方法。首先利用YOLOv4网络检测出车辆及行人在图像中的位置,之后基于坐标升维逆映射将检测出的物体位置升维映射至世界坐标系下,最后将三维模型渲染在相应的逆映射位置上来代替显示畸变的三维物体,从而给驾驶员提供有效的周边物体位置信息。实验结果表明,所提方法生成的增强型车辆三维全景影像具有很好的实时性和显示效果,能够有效解决当前车辆三维全景影像的显示缺陷。

关键词: 全景影像, 深度学习, 图像处理, 坐标变换, 增强现实, YOLOv4

Abstract: The current state-of-the-art vehicle 3D surround view system can realistically display the 3D surround environment of the vehicle body, but it still causes display distortion of the 3D objects close to the vehicle body, greatly decreasing the display effect and the practicality. To solve this problem, an enhanced vehicle 3D surround view synthesis method was proposed. First, the You Only Look Once v4(YOLOv4) network was used to detect the positions of the vehicles and pedestrians in images. Then, based on the coordinate dimension-increasing inverse mapping, the positions of the detected objects were mapped to the world coordinate system with dimension increased. Finally, the 3D models were placed and rendered on the corresponding inverse mapping positions to replace 3D objects with distortion, so as to provide effective position information of the surround objects. Experimental results show that the enhanced vehicle 3D surround view generated by proposed method has good real-time performance and display effect, and can effectively solve the display defects of the current vehicle 3D surround view.

Key words: surround view, deep learning, image processing, coordinate transformation, Augmented Reality (AR), You Only Look Once v4 (YOLOv4)

中图分类号: