Automatic Video Summarization Using HEVC and CNN Features

A Master of Science thesis in Computer Engineering by Obada Issa entitled, “Automatic Video Summarization Using HEVC and CNN Features”, submitted in November 2022. Thesis advisor is Dr. Tamer Shanableh. Soft copy is available (Thesis, Completion Certificate, Approval Signatures, and AUS Archives Con...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Issa, Obada (author)
التنسيق: doctoralThesis
منشور في: 2022
الموضوعات:
الوصول للمادة أونلاين:http://hdl.handle.net/11073/25177
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513444161519616
author Issa, Obada
author_facet Issa, Obada
author_role author
dc.contributor.none.fl_str_mv Shanableh, Tamer
dc.creator.none.fl_str_mv Issa, Obada
dc.date.none.fl_str_mv 2022-11
2023-03-02T07:48:03Z
2023-03-02T07:48:03Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv 35.232-2022.58
http://hdl.handle.net/11073/25177
dc.language.none.fl_str_mv en_US
dc.subject.none.fl_str_mv Static Video Summarization
Convolution Neural Networks (CNN)
Duplicate Frames
Sparse Autoencoders
Random Forests classifier
Video Coding
High Efficiency Video Codec (HEVC)
Motion estimation
Motion compensation
Stepwise regression
dc.title.none.fl_str_mv Automatic Video Summarization Using HEVC and CNN Features
dc.type.none.fl_str_mv info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/doctoralThesis
description A Master of Science thesis in Computer Engineering by Obada Issa entitled, “Automatic Video Summarization Using HEVC and CNN Features”, submitted in November 2022. Thesis advisor is Dr. Tamer Shanableh. Soft copy is available (Thesis, Completion Certificate, Approval Signatures, and AUS Archives Consent Form).
format doctoralThesis
id aus_a05dec13887a15e9dd83508e4c5058a5
identifier_str_mv 35.232-2022.58
language_invalid_str_mv en_US
network_acronym_str aus
network_name_str aus
oai_identifier_str oai:repository.aus.edu:11073/25177
publishDate 2022
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Automatic Video Summarization Using HEVC and CNN FeaturesIssa, ObadaStatic Video SummarizationConvolution Neural Networks (CNN)Duplicate FramesSparse AutoencodersRandom Forests classifierVideo CodingHigh Efficiency Video Codec (HEVC)Motion estimationMotion compensationStepwise regressionA Master of Science thesis in Computer Engineering by Obada Issa entitled, “Automatic Video Summarization Using HEVC and CNN Features”, submitted in November 2022. Thesis advisor is Dr. Tamer Shanableh. Soft copy is available (Thesis, Completion Certificate, Approval Signatures, and AUS Archives Consent Form).With the incredible surge of the internet and surveillance footage, there is a vast number of digital videos. The need to summarize these videos within databases is very crucial. This is where video summarization comes in handy. Video summarization can be achieved by a number of techniques. This study proposes a novel solution for the detection of key-frames for static video summarization. We preprocessed the well-known video datasets by coding them using the HEVC video coding standard. During coding, 64 proposed features were generated from the coder for each frame. Additionally, we extracted RGB frames from the original raw videos and fed them into pre-trained CNN networks for feature extraction. These include GoogleNet, AlexNet, Inception-ResNet-v2, and VGG16. The modified datasets are made publicly available to the research community. A subset of the proposed HEVC feature set was used to identify duplicate or similar frames and eliminate them from the video. We also propose an elimination solution based on the sum of the absolute differences between a frame and its motion-compensated predecessor. The proposed solutions are compared with existing works based on an SIFT flow algorithm that uses CNN features. Subsequently, an optional dimensionality reduction based on stepwise regression was applied to the feature vectors prior to detecting key-frames. The proposed solution is compared with existing studies that use sparse autoencoders with CNN features for dimensionality reduction. The accuracy of the proposed key-frame detection system was assessed using the Positive Predictive Values, Sensitivity, and F-score metrics. Combining the proposed solution with Multi-CNN features and using a Random Forests classifier, it was shown that the proposed solution achieved an average F-score of 0.98.College of EngineeringDepartment of Computer Science and EngineeringMaster of Science in Computer Engineering (MSCoE)Shanableh, Tamer2023-03-02T07:48:03Z2023-03-02T07:48:03Z2022-11info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/doctoralThesisapplication/pdf35.232-2022.58http://hdl.handle.net/11073/25177en_USoai:repository.aus.edu:11073/251772025-06-26T12:34:34Z
spellingShingle Automatic Video Summarization Using HEVC and CNN Features
Issa, Obada
Static Video Summarization
Convolution Neural Networks (CNN)
Duplicate Frames
Sparse Autoencoders
Random Forests classifier
Video Coding
High Efficiency Video Codec (HEVC)
Motion estimation
Motion compensation
Stepwise regression
status_str publishedVersion
title Automatic Video Summarization Using HEVC and CNN Features
title_full Automatic Video Summarization Using HEVC and CNN Features
title_fullStr Automatic Video Summarization Using HEVC and CNN Features
title_full_unstemmed Automatic Video Summarization Using HEVC and CNN Features
title_short Automatic Video Summarization Using HEVC and CNN Features
title_sort Automatic Video Summarization Using HEVC and CNN Features
topic Static Video Summarization
Convolution Neural Networks (CNN)
Duplicate Frames
Sparse Autoencoders
Random Forests classifier
Video Coding
High Efficiency Video Codec (HEVC)
Motion estimation
Motion compensation
Stepwise regression
url http://hdl.handle.net/11073/25177