In this work we introduce Lean Point Networks (LPNs) to train deeper and more accurate point processing networks by relying on three novel point processing blocks that improve memory consumption, inference time, and accuracy: a convolution-type block for point sets that blends neighborhood information in a memory-efficient manner; a crosslink block that efficiently shares information across low- and high-resolution processing branches; and a multiresolution point cloud processing block for faster diffusion of information. By combining these blocks, we design wider and deeper point-based architectures. We report systematic accuracy and memory consumption improvements on multiple publicly available segmentation tasks by using our generic modules as drop-in replacements for the blocks of multiple architectures (PointNet++, DGCNN, SpiderNet, PointCNN).
翻译:在这项工作中,我们引入了利昂点网络(LPNs)来培训更深和更准确的点处理网络,方法是依靠三个新的点处理区块来改进内存消耗、推断时间和准确性:一个用于点集的革命型区块,以内存效率的方式将周边信息混合在一起;一个交叉链接区块,在低分辨率和高分辨率处理区块之间有效分享信息;一个多分辨率点云处理区块,以更快地传播信息。我们通过结合这些区块,设计了更广泛和更深点的点基结构。我们通过使用我们通用的模块作为多个建筑区块(PointNet++、DGCNN、蜘蛛网、PointCNN)的投放替换,报告多个公开的区块的系统准确性和记忆消耗改善情况。