PIFNet: 3D Object Detection Using Joint Image and Point Cloud Features for Autonomous Driving

• Published in: Applied sciences Vol. 12; no. 7; p. 3686
• Main Authors: Zheng, Wenqi, Xie, Han, Chen, Yunfan, Roh, Jeongjin, Shin, Hyunchul
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.04.2022
• Subjects: 3D object detection; camera images; Cameras; Deep learning; lidar point cloud; Methods; Neural networks; object detection; Proposals; Semantics; Sensors; Sparsity

Owing to its wide range of applications, 3D object detection has attracted increasing attention in computer vision tasks. Most existing 3D object detection methods are based on Lidar point cloud data. However, these methods have some limitations in localization consistency and classification confide...