PROJECT TITLE :
Structured sparse representation of residue in screen content video coding
A unique rework methodology providing a regionally structured sparse model of residue for a video coding is proposed. Conventional learning-primarily based video coding strategies think about solely the amount of reworked coefficients in an exceedingly dictionary construction but disregard any structural kind of residue. A structural form of coefficients is employed for another issue, specified by a set of native block patterns, during a rework style thus as to yield the regionally compact distribution of the coefficients. A remodel coefficient coding method is additionally developed as a result of the distribution of the coefficients is different with that of typical transforms. It is demonstrated with experiments that the proposed technique outperforms the coding efficiency and visual quality over high potency video coding/screen content coding reference software.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here