Improved prediction methods for low complexity, high quality video coding
Başlık çevirisi mevcut değil.
- Tez No: 400769
- Danışmanlar: PROF. MONCEF GABBOUJ
- Tez Türü: Doktora
- Konular: Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Computer Engineering and Computer Science and Control
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2010
- Dil: İngilizce
- Üniversite: Tampereen Teknillinen Yliopisto (Tampere University of Technology)
- Enstitü: Yurtdışı Enstitü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 163
Özet
Özet yok.
Özet (Çeviri)
Video coders are an essential part of any digital video application, such as digital TV broadcasting or video conferencing as the bitrate of an uncompressed video signal is too high for any practical usage. The goal of all video coders is to maximize the video quality while minimizing the bitrate, with as low complexity as possible. This is achieved by exploiting different redundancies present in the video signal. One type of redundancy is present between different pictures captured either at different time instants from the same camera (temporal redundancy) or from different cameras at the same time instant (view redundancy). The focus of this thesis is to develop algorithms to exploit the redundancies between different pictures more effectively with the goals of improving coding efficiency of video coders and decreasing their implementation complexity. The first part of this thesis presents a detailed complexity analysis of motion compensated prediction part of the most recent H.264/AVC standard that is used to exploit the temporal redundancy. More specifically, the interpolation function of the standard is described in detail and its implementation complexity is analyzed. This part presents a novel algorithm to generate easy to decode bitstreams with high coding efficiency utilizing the flexibility of the H.264/AVC standard. The second part of this thesis discusses adaptive interpolation filtering (AIF) to improve the coding efficiency of video coders. In AIF, the coefficients of the interpolation filter are updated dynamically to compensate the time variant nature of the aliasing present in the video signal. In this part, firstly novel algorithms are presented to improve the coding efficiency of AIF schemes. This is achieved by making the filter structure and the filter symmetry more flexible so that the AIF tool becomes useful for a larger set of sequences exhibiting different statistical behavior and at different bitrates. Secondly, novel algorithms are presented to reduce both the encoding and decoding complexity of AIF schemes. This is done by using a novel structure that uses less number of taps than the state-of-art schemes, enabling an efficient implementation using 16-bit integer arithmetic. In addition, novel algorithms are developed to reduce the encoding complexity while maintaining the coding efficiency gain. The third and the final part of this thesis discusses three dimensional video coding, and specifically the Multiview Video Coding (MVC) extension of the H.264/AVC standard. MVC is a recent addition to the H.264/AVC family of video coding standards, to enable the emerging 3D video applications. The focus of this part is on the implementation complexity of MVC, paying special attention to parallel decoding of multiple views on different processors or cores. A novel algorithm is presented that encodes the views with certain restrictions so that the views could be decoded in parallel.
Benzer Tezler
- Computational aesthetics using machine learning for video game camera direction
Video oyunu kamera yönetimi için makine ögrenmesi ile hesaplamalı estetik
ALİ NACİ ERDEM
Yüksek Lisans
İngilizce
2015
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolOrta Doğu Teknik ÜniversitesiModelleme ve Simülasyon Ana Bilim Dalı
PROF. DR. UĞUR HALICI
- Düşük bir hızlarında konuşma kodlama ve uygulamaları
Low bit rate speech coding and applications
TARIK AŞKIN
Doktora
Türkçe
1999
Elektrik ve Elektronik Mühendisliğiİstanbul Teknik ÜniversitesiPROF.DR. GÜNSEL DURUSOY
- Deep learning based road segmentation from multi-source and multi-scale data
Çok kaynaklı ve çok ölçekli veriyle derin öğrenme tabanlı yol bölütlenmesi
OZAN ÖZTÜRK
Doktora
İngilizce
2023
Jeodezi ve Fotogrametriİstanbul Teknik ÜniversitesiGeomatik Mühendisliği Ana Bilim Dalı
PROF. DR. DURSUN ZAFER ŞEKER
- Nesneye yönelik sistemlerde kusurlu sınıfların öngörülmesi için makine öğrenmesi temelli bir yöntem oluşturulması
Creating a machine learning based method for predicting defective classes in object oriented systems
FİKRET AKTAŞ
Yüksek Lisans
Türkçe
2018
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiBilgisayar Mühendisliği Ana Bilim Dalı
DOÇ. DR. FEZA BUZLUCA
- Novel centrality, topology and hierarchical-aware link prediction in dynamic networks
Dinamik ağlarda merkezilik, topoloji ve hiyerarşik tabanlı bağlanti tahmini
ABUBAKHARI SSERWADDA
Doktora
İngilizce
2023
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiBilgisayar Mühendisliği Ana Bilim Dalı
DOÇ. DR. YUSUF YASLAN
YRD. DOÇ. ALPER ÖZCAN