Published on Wed Jul 01 2020

Optimisation of the PointPillars network for 3D object detection in point clouds

Joanna Stanisz, Konrad Lis, Tomasz Kryjak, Marek Gorgon

The PointPillars network offers a reasonable compromise between detection accuracy and calculation complexity. The aim of the work was to propose a variant of the network which we will ultimately implement in an FPGA device.

0
0
0
Abstract

In this paper we present our research on the optimisation of a deep neural network for 3D object detection in a point cloud. Techniques like quantisation and pruning available in the Brevitas and PyTorch tools were used. We performed the experiments for the PointPillars network, which offers a reasonable compromise between detection accuracy and calculation complexity. The aim of this work was to propose a variant of the network which we will ultimately implement in an FPGA device. This will allow for real-time LiDAR data processing with low energy consumption. The obtained results indicate that even a significant quantisation from 32-bit floating point to 2-bit integer in the main part of the algorithm, results in 5%-9% decrease of the detection accuracy, while allowing for almost a 16-fold reduction in size of the model.

Mon Jun 10 2019
Computer Vision
Fast Hierarchical Neural Network for Feature Learning on Point Cloud
3D point clouds are intrinsically irregular and the points are sparsely distributed in a non-Euclidean space. In order to balance model performance and complexity, we introduce a novel neural network architecture exploiting local features from a manually subsampled point set. The architecture has been tested for both shape classification and segmentation.
0
0
0
Wed Sep 21 2016
Neural Networks
Vote3Deep: Fast Object Detection in 3D Point Clouds Using Efficient Convolutional Neural Networks
This paper proposes a computationally efficient approach to detecting objects in 3D point clouds using convolutional neural networks (CNNs) In particular, this is achieved by leveraging a feature-centric voting scheme to exploit the sparsity in the input.
0
0
0
Fri Dec 27 2019
Computer Vision
Deep Learning for 3D Point Clouds: A Survey
Point cloud learning has lately attracted increasing attention due to its wide applications. This paper presents a comprehensive review of recent progress in deep learning methods for point clouds. It covers three major tasks, including 3Dshape classification, 3D object detection and tracking, and 3D point cloud segmentation.
1
0
0
Wed Aug 14 2019
Machine Learning
Justlookup: One Millisecond Deep Feature Extraction for Point Clouds By Lookup Tables
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.
0
0
0
Fri Dec 14 2018
Machine Learning
PointPillars: Fast Encoders for Object Detection from Point Clouds
Object detection in point clouds is an important aspect of many robotics applications such as autonomous driving. In this work we propose PointPillars, a novel encoder which utilizes PointNets to learn a representation of point clouds organized in vertical columns (pillars)
0
0
0
Fri Jan 17 2020
Computer Vision
Review: deep learning on 3D point clouds
Point cloud has become one of the most significant data format for 3D representation. The unstructuredness of point clouds makes use of deep learning for its processing very challenging. Earlier approaches overcome this challenge by preprocessing the point cloud into a structured grid format.
0
0
0