基于點云補全的三維目標檢測
2023年電子技術應用第8期
陳輝,王帥杰,蔡晗
(桂林電子科技大學 信息與通信學院, 廣西 桂林 541004)
摘要: LiDAR技術的發展為自動駕駛提供了豐富的3D數據。然而,由于遮擋和某些反射材料的原因引起信號丟失,LiDAR點云實際上是不完整的2.5D數據,這對 3D 感知提出了根本性挑戰。針對這一問題,提出對原始數據進行三維補全的方法。根據大多數物體形狀對稱且重復率高的特點,通過學習先驗對象形狀的方法估計點云中遮擋部分的完整形狀。該方法首先識別被遮擋和信號缺失影響的區域,在這些區域中預測區域所包含對象形狀的占用概率。針對物體間遮擋的情況,通過形狀的占用概率和共享同類形狀形態進行三維補全。對自身遮擋的物體,通過自身鏡像進行恢復。最后通過點云目標檢測網絡進行學習。結果表明,通過該方法能有效地提高生成點云3D邊框的mAP(mean Average Precision)。
中圖分類號:TP389.1
文獻標志碼:A
DOI: 10.16157/j.issn.0258-7998.223624
中文引用格式: 陳輝,王帥杰,蔡晗. 基于點云補全的三維目標檢測[J]. 電子技術應用,2023,49(8):1-6.
英文引用格式: Chen Hui,Wang Shuaijie,Cai Han. 3D object detection based on point cloud completion[J]. Application of Electronic Technique,2023,49(8):1-6.
文獻標志碼:A
DOI: 10.16157/j.issn.0258-7998.223624
中文引用格式: 陳輝,王帥杰,蔡晗. 基于點云補全的三維目標檢測[J]. 電子技術應用,2023,49(8):1-6.
英文引用格式: Chen Hui,Wang Shuaijie,Cai Han. 3D object detection based on point cloud completion[J]. Application of Electronic Technique,2023,49(8):1-6.
3D object detection based on point cloud completion
Chen Hui,Wang Shuaijie,Cai Han
(School of lnformation and Communication, Guilin University of Electronic Technology, Guilin 541004, China)
Abstract: The development of LiDAR technology provides abundant 3D data for autonomous driving. However, LIDAR point cloud is actually incomplete 2.5D data due to signal loss caused by occlusion and some reflective materials, which poses a fundamental challenge to 3D perception. To solve this problem, this paper proposes a method for 3D completion of the original data. According to the symmetric shape and high repetition rate of most objects, the complete shape of the occluded part in the point cloud is estimated by learning the prior object shape. The method first identifies regions affected by occlusions and signal loss, and in these regions, predicts the occupancy probability of the shapes of objects contained in the regions. For the case of occlusion between objects, 3D completion is performed through the occupancy probability of the shape and the morphologies that share the same shape. The objects occluded by themselves are restored by mirroring themselves. Finally, it is learned through the point cloud target detection network. The results show that this method can effectively improve the mAP for generating point cloud 3D borders.
Key words : LiDAR;point cloud;3D completion;target detection
0 引言
3D目標檢測作為自動駕駛感知系統的核心基礎之一,可以廣泛應用于路徑規劃、運動預測、碰撞避免等功能。通常,帶有相應3D激光雷傳感器的汽車已經成自動駕駛領域的標準配置,由此能夠提供準確的深度信息,點云數據的處理也越來越普遍、越來越重要。盡管已有很多進展,但由于點云本質上的高度稀疏性和不規則的特性,使得傳統的卷積神經網絡無法對點云數據進行準確的學習,而且由于相機視圖和激光雷達鳥瞰視圖之間的不對齊而導致的導致模態協同和遠距離尺度變化等原因,三維點云的處理遠比二維圖像要難得多。因此,在三維點云上的目標檢測目前仍處于初級階段。
本文詳細內容請下載:http://www.rjjo.cn/resource/share/2000005474
作者信息:
陳輝,王帥杰,蔡晗
(桂林電子科技大學 信息與通信學院, 廣西 桂林 541004)
此內容為AET網站原創,未經授權禁止轉載。