Point clouds analysis with extreme learning machine and high order point CNN

The thesis presents two novel algorithms for point clouds analysis. One algorithm is Point Clouds-based Extreme Learning Machine Auto-Encoder (PC-ELM-AE) which is the first network to incorporate Extreme Learning Machine (ELM) theory for processing point clouds. PC-ELM-AE stacks multiple local struc...

全面介紹

Saved in:
書目詳細資料
主要作者: Xu, Rui
其他作者: Huang Guangbin
格式: Thesis-Master by Research
語言:English
出版: Nanyang Technological University 2020
主題:
在線閱讀:https://hdl.handle.net/10356/137132
標簽: 添加標簽
沒有標簽, 成為第一個標記此記錄!
機構: Nanyang Technological University
語言: English
id sg-ntu-dr.10356-137132
record_format dspace
spelling sg-ntu-dr.10356-1371322023-07-04T17:15:43Z Point clouds analysis with extreme learning machine and high order point CNN Xu, Rui Huang Guangbin School of Electrical and Electronic Engineering egbhuang@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Electrical and electronic engineering::Control and instrumentation::Robotics The thesis presents two novel algorithms for point clouds analysis. One algorithm is Point Clouds-based Extreme Learning Machine Auto-Encoder (PC-ELM-AE) which is the first network to incorporate Extreme Learning Machine (ELM) theory for processing point clouds. PC-ELM-AE stacks multiple local structure modules to inspect local structures and uses three ELM Auto-Encoders (ELM-AEs) to extract per-point features in hierarchy. This algorithm tremendously reduces time cost compared to deep learning methods and achieves decent performance on ModelNet40 and ModelNet10. The other algorithm is High Order Point CNN (HOPCNN) that is designed to generalize convolutions to unordered point clouds. HOPCNN combines bilinear pooling and trilinear pooling with convolutions to make convolutional outputs permutation-invariant and constrains ranks of convolutional kernels for efficient computation. HOPCNN can simultaneously benefit from high order statistics captured by bi/trilinear poolings and expressive capabilities of convolution operators. The algorithm achieves comparable or better performance than state-of-the-art on ModelNet40 and ShapeNetPart. Master of Engineering 2020-03-02T02:37:01Z 2020-03-02T02:37:01Z 2020 Thesis-Master by Research Xu, R. (2020). Point clouds analysis with extreme learning machine and high order point CNN. Master's thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/137132 10.32657/10356/137132 en This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
Engineering::Electrical and electronic engineering::Control and instrumentation::Robotics
spellingShingle Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
Engineering::Electrical and electronic engineering::Control and instrumentation::Robotics
Xu, Rui
Point clouds analysis with extreme learning machine and high order point CNN
description The thesis presents two novel algorithms for point clouds analysis. One algorithm is Point Clouds-based Extreme Learning Machine Auto-Encoder (PC-ELM-AE) which is the first network to incorporate Extreme Learning Machine (ELM) theory for processing point clouds. PC-ELM-AE stacks multiple local structure modules to inspect local structures and uses three ELM Auto-Encoders (ELM-AEs) to extract per-point features in hierarchy. This algorithm tremendously reduces time cost compared to deep learning methods and achieves decent performance on ModelNet40 and ModelNet10. The other algorithm is High Order Point CNN (HOPCNN) that is designed to generalize convolutions to unordered point clouds. HOPCNN combines bilinear pooling and trilinear pooling with convolutions to make convolutional outputs permutation-invariant and constrains ranks of convolutional kernels for efficient computation. HOPCNN can simultaneously benefit from high order statistics captured by bi/trilinear poolings and expressive capabilities of convolution operators. The algorithm achieves comparable or better performance than state-of-the-art on ModelNet40 and ShapeNetPart.
author2 Huang Guangbin
author_facet Huang Guangbin
Xu, Rui
format Thesis-Master by Research
author Xu, Rui
author_sort Xu, Rui
title Point clouds analysis with extreme learning machine and high order point CNN
title_short Point clouds analysis with extreme learning machine and high order point CNN
title_full Point clouds analysis with extreme learning machine and high order point CNN
title_fullStr Point clouds analysis with extreme learning machine and high order point CNN
title_full_unstemmed Point clouds analysis with extreme learning machine and high order point CNN
title_sort point clouds analysis with extreme learning machine and high order point cnn
publisher Nanyang Technological University
publishDate 2020
url https://hdl.handle.net/10356/137132
_version_ 1772828191582322688