MARC 主機 00000nam a2200457K 4500 001 AAI10743336 005 20180608102941.5 006 m o u 007 cr mn||||a|a|| 008 180608s2017 xx sbm 000 0 eng d 020 9780355615777 035 (MiAaPQ)AAI10743336 035 (MiAaPQ)umkc:11232 040 MiAaPQ|beng|cMiAaPQ 100 1 Patel, Marmikkumar 245 10 H3DNET :|bA Deep Learning Framework for Hierarchical 3D Object Classification 264 0 |c2017 300 1 online resource (97 pages) 336 text|btxt|2rdacontent 337 computer|bc|2rdamedia 338 online resource|bcr|2rdacarrier 500 Source: Masters Abstracts International, Volume: 57-04 500 Adviser: Yugyung Lee 502 Thesis (M.S.)--University of Missouri - Kansas City, 2017 504 Includes bibliographical references 520 Deep learning has received a lot of attention in the fields such as speech recognition and image classification because of the ability to learn multiple levels of features from raw data. However, 3D deep learning is relatively new but in high demand with their great research values. Current research and usage of deep learning for 3D data suffer from the limited ability to process large volumes of data as well as low performance, especially in increasing the number of classes in the image classification task. One of the open questions is whether an efficient, as well as an accurate 3D Deep Learning model, can be built with large-scale 3D data 520 In this thesis, we aim to design a hierarchical framework for 3D Deep Learning, called H3DNET, which can build a DL 3D model in a distributed and scalable manner. In the H3DNET framework, a learning problem is composed of two stages: divide and conquer. At the divide learning stage, a learning problem is divided into several smaller problems. At the conquer learning stage, an optimized solution is used to solve these smaller subproblems for a better learning performance. This involves training of models and optimizing them with the refined division for a better performance. The inferencing can achieve the efficiency and high accuracy with fuzzy classification using such a two-step approach in a hierarchical manner 520 The H3DNET framework was implemented in TensorFlow which is capable of using GPU computations in parallel to build the 3D neural network. We evaluated the H3DNET framework on a 3D object classification with MODELNET10 and MODELNET40 datasets to check the efficiency of the framework. The evaluation results verified that the H3DNET framework supports hierarchical 3D Deep Learning with 3D images in a scalable manner. The classification accuracy is higher than the state-of-the-art, VOXNET and POINTNET 533 Electronic reproduction.|bAnn Arbor, Mich. :|cProQuest, |d2018 538 Mode of access: World Wide Web 650 4 Computer science 650 4 Artificial intelligence 650 4 Library science 655 7 Electronic books.|2local 690 0984 690 0800 690 0399 710 2 ProQuest Information and Learning Co 710 2 University of Missouri - Kansas City.|bComputer Science 773 0 |tMasters Abstracts International|g57-04(E) 856 40 |uhttp://pqdd.sinica.edu.tw/twdaoapp/servlet/ advanced?query=10743336|zclick for full text (PQDT) 912 PQDT
館藏地 | 索書號 | 條碼 | 處理狀態 |
---|