IPOD: Intensive Point-based Object Detector for Point Cloud

13 Dec 2018  ·  Zetong Yang, Yanan sun, Shu Liu, Xiaoyong Shen, Jiaya Jia ·

We present a novel 3D object detection framework, named IPOD, based on raw point cloud. It seeds object proposal for each point, which is the basic element... This paradigm provides us with high recall and high fidelity of information, leading to a suitable way to process point cloud data. We design an end-to-end trainable architecture, where features of all points within a proposal are extracted from the backbone network and achieve a proposal feature for final bounding inference. These features with both context information and precise point cloud coordinates yield improved performance. We conduct experiments on KITTI dataset, evaluating our performance in terms of 3D object detection, Bird's Eye View (BEV) detection and 2D object detection. Our method accomplishes new state-of-the-art , showing great advantage on the hard set. read more

PDF Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
3D Object Detection KITTI Cars Easy IPOD AP 79.75% # 17
3D Object Detection KITTI Cars Hard IPOD AP 66.33% # 14
3D Object Detection KITTI Cars Moderate IPOD AP 72.57% # 16
3D Object Detection KITTI Cyclists Easy IPOD AP 71.40% # 8
3D Object Detection KITTI Cyclists Hard IPOD AP 48.34% # 8
3D Object Detection KITTI Cyclists Moderate IPOD AP 53.46% # 8
3D Object Detection KITTI Pedestrians Easy IPOD AP 56.92% # 1
3D Object Detection KITTI Pedestrians Hard IPOD AP 42.39% # 2
3D Object Detection KITTI Pedestrians Moderate IPOD AP 44.68% # 3

Methods


No methods listed for this paper. Add relevant methods here