P2Net: A Post-Processing Network for Refining Semantic Segmentation of LiDAR Point Cloud based on Consistency of Consecutive Frames
P2Net: A Post-Processing Network for Refining Semantic Segmentation of LiDAR Point Cloud based on Consistency of Consecutive Frames
复制标题
DOI：
10.1109/smc42975.2020.9283329
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Yutaka Momma;Weimin Wang;E. Simo-Serra;S. Iizuka;R. Nakamura;H. Ishikawa
中科院分区：
文献类型：
--
作者：
Yutaka Momma;Weimin Wang;E. Simo-Serra;S. Iizuka;R. Nakamura;H. Ishikawa
我们提出了一种轻量级后处理方法来细化点云序列的语义分割结果。大多数现有方法通常逐帧分割，并遇到问题固有的模糊性：基于单帧中的测量，标签有时甚至对于人类来说也难以预测。为了解决这个问题，我们建议显式地训练一个网络来细化现有分割方法预测的这些结果。该网络（我们称之为 P2Net）在注册后从连续帧中学习“重合”点之间的一致性约束。我们在由真实户外场景组成的 SemanticKITTI 数据集上定性和定量地评估了所提出的后处理方法。通过比较两个代表性网络在经过后处理网络和未经后处理网络的细化后的预测结果，验证了该方法的有效性。具体来说，定性可视化验证了以下关键思想：难以预测的点的标签可以通过 P2Net 进行纠正。从数量上看，PointNet [1] 的总体 mIoU 从 10.5% 提高到 11.7%，PointNet++ [2] 从 10.8% 提高到 15.9%。
We present a lightweight post-processing method to refine the semantic segmentation results of point cloud sequences. Most existing methods usually segment frame by frame and encounter the inherent ambiguity of the problem: based on a measurement in a single frame, labels are sometimes difficult to predict even for humans. To remedy this problem, we propose to explicitly train a network to refine these results predicted by an existing segmentation method. The network, which we call the P2Net, learns the consistency constraints between "coincident" points from consecutive frames after registration. We evaluate the proposed post-processing method both qualitatively and quantitatively on the SemanticKITTI dataset that consists of real outdoor scenes. The effectiveness of the proposed method is validated by comparing the results predicted by two representative networks with and without the refinement by the post-processing network. Specifically, qualitative visualization validates the key idea that labels of the points that are difficult to predict can be corrected with P2Net. Quantitatively, overall mIoU is improved from 10.5% to 11.7% for PointNet [1] and from 10.8% to 15.9% for PointNet++ [2].