PointPWC-Net: Cost Volume on Point Clouds for (Self-)Supervised Scene Flow Estimation
PointPWC-Net: Cost Volume on Point Clouds for (Self-)Supervised Scene Flow Estimation
复制标题
DOI：
10.1007/978-3-030-58558-7_6
复制
发表时间：
2020
影响因子：
6.6
通讯作者：
Wenxuan Wu;Zhiyuan Wang;Zhuwen Li;Wei Liu;Fuxin Li
中科院分区：
文献类型：
--
作者：
Wenxuan Wu;Zhiyuan Wang;Zhuwen Li;Wei Liu;Fuxin Li
我们提出了一种新的端到端的深度场景流模型，称为PointPWC-Net，它直接以粗到细的方式处理具有大运动的3D点云场景。在粗略级别计算的流被上采样并扭曲到更精细的级别，使算法能够适应大的运动，而没有禁止的搜索空间。我们引入了新的成本体积，上采样和扭曲层，以有效地处理3D点云数据。与传统的成本卷需要在高维网格上穷尽计算所有成本值不同，我们的基于点的公式将成本卷离散化到输入3D点上，PointConv操作有效地计算成本卷上的卷积。在FlyingThings 3D和KITTI上的实验结果大大优于现有技术。我们进一步探索了新的自监督损失来训练我们的模型，并取得了与最先进的监督损失训练相当的结果。在没有任何微调的情况下，我们的方法在KITTI Scene Flow 2015数据集上也表现出了很好的泛化能力，优于所有以前的方法。代码发布于 https://github.com/DylanWusee/PointPWC .
We propose a novel end-to-end deep scene flow model, called PointPWC-Net, that directly processes 3D point cloud scenes with large motions in a coarse-to-fine fashion. Flow computed at the coarse level is upsampled and warped to a finer level, enabling the algorithm to accommodate for large motion without a prohibitive search space. We introduce novel cost volume, upsampling, and warping layers to efficiently handle 3D point cloud data. Unlike traditional cost volumes that require exhaustively computing all the cost values on a high-dimensional grid, our point-based formulation discretizes the cost volume onto input 3D points, and a PointConv operation efficiently computes convolutions on the cost volume. Experiment results on FlyingThings3D and KITTI outperform the state-of-the-art by a large margin. We further explore novel self-supervised losses to train our model and achieve comparable results to state-of-the-art trained with supervised loss. Without any fine-tuning, our method also shows great generalization ability on the KITTI Scene Flow 2015 dataset, outperforming all previous methods. The code is released at https://github.com/DylanWusee/PointPWC .