Justlookup: One Millisecond Deep Feature Extraction for Point Clouds By Lookup Tables

Abstract

Deep models are capable of fitting complex high dimensional functions while usually yielding large computation load. There is no way to speed up the inference process by classical lookup tables due to the high-dimensional input and limited memory size. Recently, a novel architecture (Point-Net) for point clouds has demonstrated that it is possible to obtain a complicated deep function from a set of 3-variable functions. In this paper, we exploit this property and apply a lookup table to encode these 3-variable functions. This method ensures that the inference time is only determined by the memory access no matter how complicated the deep function is. We conduct extensive experiments on ModelNet and ShapeNet datasets and demonstrate that we can complete the inference process in 1.5 ms on an Intel i7-8700 CPU (single core mode), 32× speedup over the PointNet architecture without any performance degradation.

Deep Learning for 3D Point Clouds: A Survey

“…Duan et al [32] proposed a Structural Relational Network (SRN) to learn structural relational features between different local structures using MLP. Lin et al [33] accelerated the inference process by constructing a lookup table for both input and function spaces learned by PointNet. The inference time on the ModelNet and ShapeNet datasets is sped up by 1.5 ms and 32 times over PointNet on a moderate machine.…”

PointSCNet: Point Cloud Structure and Correlation Learning Based on Space Filling Curve-Guided Sampling

“…Then, the PointNet++ [13] proposed a multi-layer sampling and grouping method to improve the PointNet. Much research on point cloud processing [27][28][29][30][31] later followed the idea of pointwise and hierarchical point feature extraction. However, the feature extraction in PointNet ignores geometrical structure information and the potential relationship between the local regions.…”

Multi-view dual attention network for 3D object recognition

“…?, and it uses the context of local neighborhoods to use adaptive feature adjustment (AFA) to improve point features. Lin et al [24] accelerated the reasoning process by constructing a lookup table for the input and function spaces of PointNet learning. This method of directly using unordered point cloud as input has always been the pursuit of 3D object recognition.…”