论文标题
点云和事件流网络连续域上的稀疏卷积
Sparse Convolutions on Continuous Domains for Point Cloud and Event Stream Networks
论文作者
论文摘要
图像卷积一直是计算机视觉中大量深度学习进步的基石。研究界尚未确定一个等效的操作员，用于稀疏，非结构化的连续数据（例如点云和事件流）。对于这些情况，我们提供了对卷积运算符的优雅稀疏基质解释，这与训练过程中卷积和有效的数学定义一致。在基准点云分类问题上，我们演示了使用这些操作构建的网络可以比现有方法更快或更快地训练数量级或更快，同时保持可比较的精度并需要一小部分内存。我们还将操作员应用于事件流处理，从而在多个任务上实现了最先进的结果，并具有数十万个事件的流。
Image convolutions have been a cornerstone of a great number of deep learning advances in computer vision. The research community is yet to settle on an equivalent operator for sparse, unstructured continuous data like point clouds and event streams however. We present an elegant sparse matrix-based interpretation of the convolution operator for these cases, which is consistent with the mathematical definition of convolution and efficient during training. On benchmark point cloud classification problems we demonstrate networks built with these operations can train an order of magnitude or more faster than top existing methods, whilst maintaining comparable accuracy and requiring a tiny fraction of the memory. We also apply our operator to event stream processing, achieving state-of-the-art results on multiple tasks with streams of hundreds of thousands of events.
