Improved pixel prediction for video coding

Background

Pixel prediction is an important part of video coding standards such as H.261, H.263, MPEG-4 and H.264. In H.264 there are three pixel prediction methods utilized namely intra, inter and bi-prediction. Intra prediction provides a spatial prediction of the current block from previously decoded pixels of the current frame. Inter prediction gives a temporal prediction of the current block using a corresponding but displaced block in a previously decoded frame. Bi-directional prediction gives a weighted average of two inter predictions. The difference between the original block and the predicted block is then coded by a transform to remove remaining redundancies.

Task

The goal with this thesis is to study and implement approaches to improve the coding efficiency by improving the pixel prediction mechanisms as in H.264. This work includes topics as filtering, video coding and human visual perception.

The main work will be done on top of existing H.264 encoder/decoder software (C code) developed at Ericsson Research.

The thesis will contain:
• Theoretical studies
• Implementation of algorithms in software
• Objective and subjective evaluation of algorithms with respect to compression efficiency
• Technical report

Qualifications

The student is expected to have fulfilled the course requirements of a masters degree. We are looking for a student with interest and knowledge in algorithm design. Good skills in programming in C and writing in English are also required.

Ericsson Research – Multimedia technologies

The thesis work will be done at the Visual Technology section of Multimedia Technologies at Ericsson Research in Kista, Torshamnsgatan 23.

Contact:
Kenneth Andersson, [email protected], 08-764 15 38

Applications (containing a short personal letter, recent grades and CV) should be sent by e-mail to: [email protected]

Please apply before: 2007-09-15