Search or add a thesis

Advanced Search (Beta)
Home > Fast Video Encoding Using Spatio-Temporal Features and Ensemble Classification

Fast Video Encoding Using Spatio-Temporal Features and Ensemble Classification

Thesis Info

Access Option

External Link

Author

Tahir, Muhammad

Program

PhD

Institute

Capital University of Science & Technology

City

Islamabad

Province

Islamabad.

Country

Pakistan

Thesis Completing Year

2019

Thesis Completion Status

Completed

Subject

Electrical Engineering

Language

English

Link

http://prr.hec.gov.pk/jspui/bitstream/123456789/11801/1/Muhammad%20Tahir%20EE%202019%20cust%20isb%20prr.pdf

Added

2021-02-17 19:49:13

Modified

2024-03-24 20:25:49

ARI ID

1676727757893

Similar


Video Coding has evolved over the years and new compression standards are being developed at regular intervals. Latest video codecs such as High Efficiency Video Coding (HEVC) have improved compression efficiency to reduce the bit-rates of en coded streams. This improvement has resulted in high computational complexity that becomes a bottleneck in real-time implementation of these codecs. Reduction in this computational complexity without compromise on the video quality is a challenge. Another challenge in fast video encoding is the diverse nature of video content. Hence, there is need for development of intelligent techniques to reduce the computational complexity of latest video codecs that also adapt to the diverse nature of video data. Research in this thesis focuses on identification of local and global features ex tracted from video data that can be used to characterize the diverse content in video sequences. Texture variations and motion content in video sequence are quantified to categorize it into simple and complex video sequence. This informa tion is used to develop a content adaptive fast encoding framework to reduce the computational complexity of HEVC. Proposed framework performs equally well both for sequences with simple or complex video content without compromising on video quality. It has been tested with a large set of video sequences and shows promising results as compared to the other recent works. This research work also focuses on the use of machine learning based algorithms for fast encoding to significantly reduce the complexity of video codecs while keeping bit-rate and PSNR within limits in recent video standards like HEVC. Machine learning based techniques formulate encoding process as a classification problem and use features extracted from video data to model the classifiers that can assist in early predictions during encoding. A large set of spatial and temporal features is extracted from video data and a systematic approach is applied for optimal feature selection. Resultant optimal features are used to train a Random Forests based ensemble classifier for early selection of prediction modes and coding unit sizes in HEVC. Proposed technique has been evaluated with publicly available video data x sets. Experimental results show that the proposed approach significantly reduces the complexity of different profiles in HEVC without compromising on video qual ity and performs better than other existing fast video encoding implementations. Fast encoding methods developed in this research have also been validated using emerging applications of HEVC such as compression of light field images. Detailed analysis of different formats in light field image representation has been carried out to identify unique aspects that differentiate them from natural videos. These unique features are used in fast coding of light field images in various formats using HEVC. Experimental results show that proposed technique can be applied in fast coding of light field images without compromise on image quality. These results can be used as benchmark for future research on fast encoding of light field images. Hence, the fast video encoding methods developed in this research can be extended to other applications of image and video coding. These techniques can also be integrated in real life video encoding solutions to enable implementation of latest video codecs on embedded hardware platforms with limited processing power and memory.
Loading...
Loading...

Similar Books

Loading...

Similar Chapters

Loading...

Similar News

Loading...

Similar Articles

Loading...

Similar Article Headings

Loading...

عائلی زندگی

عائلی زندگی

                ناطق کی شادی 2010 میں ہوئی۔ان کی زوجہ پاک آرمی میں ڈاکٹر تھیں ۔ان کی شادی تقریباً 5 سال رہی لیکن یہ شادی مزید نہ چل سکی اور بیوی سے علیحدگی ہو گئی یعنی 2015 ئ￿ میں طلاق ہو گئی۔انہوں نے شادی کا دوبارہ تاحال نہیں سوچا۔ اللہ تعالیٰ نے بیٹی جیسی رحمت سے نوازا جس کا نام’’وجیہہ فاطمہ ‘‘ ہے۔بیٹی اپنی والدہ کے ساتھ لاہور میں ہی رہتی ہیں۔

Examining the Issue of Identity in Ayisha Malik’s Novel “Sofia Khan is Not Obliged” Through Homi K. Bhabha’s Concepts of ‘Hybridity, Ambivalence and Mimicry’

This paper studies Ayisha Malik’s Sofia Khan is not obliged from postcolonial perspective. The paper studies the novel from the view of the discourse presented by Homi K. Bhabha. The novel was published in 2015. The story revolves around a girl of Muslim ethnicity from Pakistan named Sofia Khan. She is living in London and is working there in a publishing company. The story is about the adventures of protagonist; her experiences and views the London from her perspective. The paper analyzes the characters and the main events from the research method of textual analysis. The study finds that the concept proposed by Bhabha, very much plays an operative role (viable role) on the British-Muslim- characters in their attempt to assimilate into their Host/Home country. It concludes that how these Muslim characters locate agency in the “in between space” within the process of mimicry and negotiate their identity in their effort to assimilate in colonizer’s space.

Coordination Compounds of Silver and Copper With S- and N-Donor Ligands for Potential Biological Application

This study deals with the complexes of copper and silver with nitrogen and sulphur donor ligands for potentional biological applications. Our main research was focused onthe synthesisof various ligands containing sulphur and nitrogen donor atoms and some commercially available ligands were also used in complexation with different metal ions such as copper and silver. In most of the cases, we got our expected products with no side reactions, for purification of organic derivatives column chromatography was extensively used. Different ligands which contained sulphur and nitrogen donor sites were synthesized and coupled with different metal salts of the above mentioned metals including pallidium and cobalt in few cases. These ligands can be divided into three different groups 1) quinoline based ligands 2) tetrahydroindazole based ligands 3) thiazole based ligands. The quinoline based ligands were synthesized by multistep reactions. All the reactions gave expected products with negligible side reactions. Compounds 1-9 in the following text are quinoline based ligands and their complexes. All the qunoline based complexes were tested for antifungal, antibacterial, antioxidant and enzymes inhibition activities. The best antibacterial agents among quinoline based complexes waspalladium based compound 7. Similarly the best antifungal agent among quinoline based complexes is compound 6. The antioxidant study showed that compound 7 hasthe highest antioxidant activity among quinoline based complexes, against both ABTS and DPPH. Similarly compounds 7 also showed highest enzymes inhibation against AChE and BChE. Two compounds 10 and 11 are ligands containing N and O donor sites, but at ordinary condition, their complexation with a suitable metal ion was not feasible. Compounds 12-16 are indazole complexes of copper and silver and they were checked for their antifungal, antibacterial, antioxidant and enzymes inhibition activities. Compound 12 was found most effective agent against E.coli, compound 13 was efficient against K. Pneumoniaboth exhibiting 16 mm zone of inhibition. Compounds 12 and 13 showed good activity against A. fumegatus. The antioxidant test of the indazole based complexes showed that compound 15 had better anti-oxidant activity, up to 73% (IC50 value 94µg/mL) and was most effective antioxidant among compounds of the series. Similarly in thiazole complexes17-20, the most active against E.coli and K. Pneumonia was compound 17 which exhibited 19 and 14 mm zone of inhibition, respectively. The antifungal test of the thiazole compounds showed that compounds 18 and 19 were comparatively more active against A. fumegatusas compared to other members of the series. In thiazole complexes 18 was better anti-oxidant agent, exhibiting IC50 value 69µg/mL for ABTS and 50µg/mL for DPPH. Compound 19 had low antioxidant activity IC50= 314µg/mL. Compound 20 was very active AChE inhibitor with 77% activity at 1000 µg/mL. Compound 18 inhibited BchEupto 74% (caculated IC50= 147 µg/mL).