Max-Planck-Institut für Informatik
max planck institut
informatik
mpii logo Minerva of the Max Planck Society
 

MPI-INF or MPI-SWS or Local Campus Event Calendar

<< Previous Entry Next Entry >> New Event Entry Edit this Entry Login to DB (to update, delete)
What and Who
Title:Hybrid Model Based Coding Enhancement for MPEG
Speaker:Sukhpreet Kaur Khangura
coming from:Saarland University
Speakers Bio:
Event Type:PhD Application Talk
Visibility:D1, D2, D3, D4, D5, SWS, RG1, MMCI
We use this to send out email in the morning.
Level:Public Audience
Language:English
Date, Time and Location
Date:Monday, 19 May 2014
Time:10:45
Duration:60 Minutes
Location:Saarbr├╝cken
Building:E1 4
Room:024
Abstract
The scope of the Master's thesis is to integrate the Model Based Coding (MBC) with the MPEG-4 (i.e. ISO/IEC 14496-5) reference software to have Hybrid Model Based Video Codec. MPEG-4 (ISO/IEC 14496-5) provides reference software for demonstrating and clarifying the ITU-T H.264/MPEG-4 (Part 10) Advanced Video Coding standard. We have modified the prediction mechanism in the MPEG-4 so that the I and P frames can be predicted from a static image-Model frame. To achieve this prediction we have integrated the proof-of-concept renderer into the MPEG-4 reference software.

The novelty of our approach lies in the fact that it helps in the fully synthetic prediction of the first frames which occur after scene change. This prediction helps in reduction of size of the frames and in total bits. We have found that the high Signal to-Noise Ratio (SNR) of the Model can result in improving the SNR of the first frame. This high SNR of first frame further helps in reducing the size of the subsequent frames. The prediction is done well when there is a less difference between the Model and the incoming first frame. We have also found that significant reduction can be achieved for the input video sequence which has scene change or complex environment by providing multiple Models corresponding to each object in the input video.

We have found that the reduction is highest when the Model and the object are at the same position and aligned in one direction. We have found that by using highly professional renderer, the reduction can be achieved for real video sequences as well. We have shown the visual effects which have indicated where the content is encoded Model based and where it is traditionally encoded. It has been found that by enhancing MPEG-4 with MBC results in the reduction of the total bits which are required for video transmission because the decoder can make use of Model information which it has already. Hence, this information is not required to be transmitted again every time. We have found that the decreased bitrate has not come at the cost of visual quality as can be seen from the Peak Signal-to-Noise Ratio (PSNR) values of luma and chroma components.

Contact
Name(s):Jennifer Gerling
Phone:1800
EMail:--email address not disclosed on the web
Video Broadcast
Video Broadcast:NoTo Location:
Tags, Category, Keywords and additional notes
Note:
Attachments, File(s):

Created:Jennifer Gerling/AG2/MPII/DE, 05/15/2014 11:36 AM Last modified:Uwe Brahm/MPII/DE, 11/24/2016 04:13 PM
  • Jennifer Gerling, 05/15/2014 11:37 AM -- Created document.