Apparatus and method for performing error concealment of inter-coded video frames

摘要

An embodiment relates to performing error concealment of a corrupted block in a video frame, which is capable of performing a real time reconstruction of corrupted blocks which allow to precisely recover small details and fine movements, in particular, the error concealment apparatus and method according to an embodiment selects a replacement block by taking into account the luminance distortion and the motion characteristics of the video sequence. The latter is represented by the distance of the motion vectors chosen as candidate replacements and the average value of the motion vectors of the blocks surrounding the missing block in the current frame.

主权项

1. An apparatus, comprising:
a first unit configured to determine a first level of distortion associated with a first region of a reference first image and with a recovered second region of a second image that is being decoded with reference to the first image, the second region being adjacent to an unrecovered third region of the second image; a second unit configured to determine a first relationship between a reference motion vector and a motion vector for the second region; and a third unit configured to replace the unrecovered third region of the second image in response to the level of distortion associated with the first and second regions and in response to the relationship between the reference motion vector and the motion vector for the second region, wherein the third unit is configured to replace the unrecovered third region of the second image with the first region of the first image in response to a sum of the level of distortion associated with the first and second regions and the relationship between the reference motion vector and the motion vector for the second region.