Video encoding
AssigneeKoninklijke Philips Electronics N.V.
InventorWilhelmus Hendrikus Alfonsus Brüls
Life of this patent
Estimating the term…
Abstract
To efficiently use space capacity and to provide accurate motion vectors for P-pictures, a two-pass search in a P-coding mode of an MPEG encoder is run so that the second pass refines the precision of motion vectors found in the first pass. Within a frame encoding period, the first pass coincides with the time interval used by a known MPEG video encoder in encoding B-pictures to search forward motion vectors with respect to a prior image, and, within the frame encoding period, the second pass coincides with searching backward motion vectors with respect to a subsequent image.
Claims6
A video encoder for predictively encoding images with reference to one reference image in a first (P) prediction mode and to two reference images in a second (B) prediction mode, comprising a motion estimation circuit using a first interval of a frame encoding period in the second (B) prediction mode to search motion vectors representing motion between an input image and one of said two reference images, and using a second interval of said frame encoding period to search motion vectors representing motion between said input image and the other one of said two reference images;
- characterized in that the motion estimation circuit is arranged to use the first interval of the frame encoding period in the first (P) prediction mode to search motion vectors representing motion between an input image and said one reference image, and to use the second interval of said frame encoding period to refine the motion vectors found in the first interval.
A method of predictively encoding images with reference to one reference image in a first (P) prediction mode and to two reference images in a second (B) prediction mode, comprising the steps of searching motion vectors representing motion between an input image and one of said two reference images in a first interval of a frame encoding period in the second (B) prediction mode, and searching motion vectors representing motion between said input image and the other one of said two reference images in a second interval of said frame encoding period, characterized by the steps of searching motion vectors representing motion between an input image and said one reference image in the first interval of the frame encoding period in the first (P) prediction mode, and refining the motion vectors found in the first interval in the second interval of said frame encoding period.
Dependencies are inferred from the claim text.