Simple linear iterative clustering based low-cost pseudo-LiDAR for 3D object detection in autonomous driving

被引:0
作者
Duy Le
Linh Nguyen
机构
[1] The Australian National University,College of Engineering & Computer Science
[2] Federation University Australia,School of Engineering, Information Technology and Physical Sciences
来源
Multimedia Tools and Applications | 2023年 / 82卷
关键词
LiDAR; 3D object detection; Simple linear iterative clustering; Superpixel; Autonomous driving;
D O I
暂无
中图分类号
学科分类号
摘要
The paper presents a low-cost and LiDAR-free approach to efficiently detect 3D objects from stereo camera images, towards autonomous driving applications. It is first proposed to exploit the simple linear iterative clustering algorithm to segment stereo images into superpixel feature maps. The segmented superpixel maps are then used to estimate a depth map. By utilizing the depth map and stereo images, a 3D point cloud can be generated; and the 3D data is considered as pseudo-LiDAR representation as it is similar to measurements collected by a LiDAR sensor. The generated pseudo-LiDAR point cloud can ultimately be fed into any the state-of-the-art LiDAR based 3D object detection techniques to localize objects. By doing this, the proposed approach can effectively detect 3D objects by only employing low-cost stereo cameras, which can save tens of thousands of dollars on LiDAR costs from the existing LiDAR based methods. Effectiveness of the proposed algorithm was evaluated in the real-world KITTI dataset where the obtained results are about 1.33% better than those obtained by the benchmarking pseudo-LiDAR++ method (You et al. 2020).
引用
收藏
页码:25253 / 25269
页数:16
相关论文
共 58 条
  • [1] Achanta R(2012)SLIC Superpixels compared to state-of-the-art superpixel methods IEEE Trans Pattern Anal Mach Intell 34 2274-2282
  • [2] Shaji A(2019)A survey on 3d object detection methods for autonomous driving applications IEEE Trans Intell Transp Syst 20 3782-3795
  • [3] Smith K(2018)Superpixel segmentation using gaussian mixture model IEEE Trans Image Process 27 4105-4117
  • [4] Lucchi A(2021)Yolov4-5d: an effective and efficient object detector for autonomous driving IEEE Trans Instrum Meas 70 1-13
  • [5] Fua P(2018)3D object proposals using stereo imagery for accurate object class detection IEEE Trans Pattern Anal Mach Intell 40 1259-1272
  • [6] Süsstrunk S(1997)A study of efficiency and accuracy in the transformation from rgb to cielab color space IEEE Trans Image Process 6 1046-1048
  • [7] Arnold E(1981)Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography Commun ACM 24 381-395
  • [8] Al-Jarrah OY(2013)Vision meets robotics: the KITTI dataset Int J Robot Res 32 1231-1237
  • [9] Dianati M(2021)Monocular 3-d vehicle detection using a cascade network for autonomous driving IEEE Trans Instrum Meas 70 1-13
  • [10] Fallah S(2002)An efficient k-means clustering algorithm: analysis and implementation IEEE Trans Pattern Anal Mach Intell 24 881-892