Method and apparatus for compensating quantization errors of a decoded video image by using an adaptive filter
US5946421A · kind A · utility
Assignee
Inventor
Key dates
| Filing date | Aug 28, 1997 |
| Grant date | Aug 31, 1999 |
| Priority date | — |
| Expiry date | Aug 28, 2017 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH04N19/80
- WIPO fieldAudio-visual technology
- WIPO sectorElectrical engineering
Abstract
An encoded video signal is decoded to obtain a decoded video signal, wherein the decoded video signal includes a previous and a current frames and is processed an a macroblock-by-macroblock basis, each frame's data containing pixel values, by deriving a set of quantized discrete cosine transform coefficients corresponding to current frame data and converting the set of quantized discrete cosine transform coefficients into a set of discrete cosine transform coefficients. Thereafter, depending on the complexity of the current frame data detected, high frequency components of the set of discrete cosine transform coefficients within a preset region are filtered based on the set of discrete cosine transform coefficients and the complexity of the current frame data, to thereby produce a set of filtered discrete cosine transform coefficients. The set of filtered discrete cosine transform coefficients is transformed into a set of inverse discrete cosine transform data, and the current frame data is generated through the use of motion compensation and provided as the decoded video signal.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.