東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back to Search results for [ null ]

Switch To: Labeled | MARC Mode | ISBD

Topic-Based Video Classification and...

Vadlamudi, Naga Krishna.

Linked to FindBook

Google Book

Amazon

博客來

Topic-Based Video Classification and Retrieval Using Machine Learning.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Topic-Based Video Classification and Retrieval Using Machine Learning./
Author:	Vadlamudi, Naga Krishna.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2017,
Description:	77 p.
Notes:	Source: Masters Abstracts International, Volume: 57-04.
Contained By:	Masters Abstracts International57-04(E).
Subject:	Computer science. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10743254
ISBN:	9780355615753

Topic-Based Video Classification and Retrieval Using Machine Learning.
Vadlamudi, Naga Krishna.

Topic-Based Video Classification and Retrieval Using Machine Learning. - Ann Arbor : ProQuest Dissertations & Theses, 2017 - 77 p.

Source: Masters Abstracts International, Volume: 57-04.

Thesis (M.S.)--University of Missouri - Kansas City, 2017.

Machine learning has made significant progress for many real-world problems. The Deep Learning (DL) models proposed primarily concentrate on object detection, image classification, and image captioning. However, very little work has been shown in DL-based video-content analysis and retrieval. Due to the complex nature of time relevant information in a sequence of video frames, understanding video contents is particularly challenging during video analysis and retrieval. Latent Dirichlet Allocation (LDA) is known as one of the best-proven methods for uncovering hidden latent semantic structures (called the topics) from a large corpus. We want to extend it to capture topics from annotated videos and effectively use them for video classification and retrieval.

ISBN: 9780355615753Subjects--Topical Terms:

523869
Computer science.

Topic-Based Video Classification and Retrieval Using Machine Learning.
LDR:02639nmm a2200325 4500 001 2157680
005 20180608102941.5
008 190424s2017 ||||||||||||||||| ||eng d
020 $a 9780355615753
035 $a (MiAaPQ)AAI10743254
035 $a (MiAaPQ)umkc:11230
035 $a AAI10743254
040 $a MiAaPQ $c MiAaPQ
100 1 $a Vadlamudi, Naga Krishna. $3 3345495
245 1 0 $a Topic-Based Video Classification and Retrieval Using Machine Learning.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2017
300 $a 77 p.
500 $a Source: Masters Abstracts International, Volume: 57-04.
500 $a Adviser: Yugyung Lee.
502 $a Thesis (M.S.)--University of Missouri - Kansas City, 2017.
520 $a Machine learning has made significant progress for many real-world problems. The Deep Learning (DL) models proposed primarily concentrate on object detection, image classification, and image captioning. However, very little work has been shown in DL-based video-content analysis and retrieval. Due to the complex nature of time relevant information in a sequence of video frames, understanding video contents is particularly challenging during video analysis and retrieval. Latent Dirichlet Allocation (LDA) is known as one of the best-proven methods for uncovering hidden latent semantic structures (called the topics) from a large corpus. We want to extend it to capture topics from annotated videos and effectively use them for video classification and retrieval.
520 $a This approach aims to classify and retrieve videos based on discovering topics from annotated keyframes in videos. This will be accomplished by employing a pipeline of the following five steps: (1) automatic keyframe detection, (2) video annotation using Show & Tell model, (3) topic discovery using LDA on the annotation, (4) topic assignment to keyframes in the videos, and (5) topic sequence analysis for videos. Mapping the topic histograms of the videos are used to both classify and retrieve videos. The unique contribution of this thesis is to design a topic histogram model that is a new way of representing topics within videos as a sequence and frequency of topics. Based on the framework, we have developed a video application using both Apache Spark and TensorFlow, and then we evaluated different machine learning algorithms and validation techniques using Wikipedia, Flickr30K, and YouTube8M datasets.
590 $a School code: 0134.
650 4 $a Computer science. $3 523869
650 4 $a Artificial intelligence. $3 516317
650 4 $a Library science. $3 539284
690 $a 0984
690 $a 0800
690 $a 0399
710 2 $a University of Missouri - Kansas City. $b Computer Science. $3 1680874
773 0 $t Masters Abstracts International $g 57-04(E).
790 $a 0134
791 $a M.S.
792 $a 2017
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10743254