Apparatus and methods for down-conversion video de-interlacing
View Patent ↗A video converter provides for converting an interlaced video datastream to a de-interlaced form suitable for progressive-input coding. Conversion is conducted in accordance with determined motion of image portions within the video datastream. In a preferred MPEG-1 and MPEG-2 IC pre-processor implementation, during motion processing, intra-frame and inter-frame motion analyses are conducted for a current-pixel. The resulting motion determinations are further correlated with derived and/or coding metrics and combined; the combination is also limited, smoothed and a transfer function is applied to produce a mix control. During further video processing, current-pixel data from a current field (“field data”) and corresponding filtered current frame data (“filtered frame data”) are mixed in accordance with the mix control. Resulting output preferably comprises filtered frame data for low or no determined motion, field data for high motion and motion-related mixtures of the two for remaining motion determinations.
1. A method for determining intra-frame motion, comprising:
receiving interlaced video data including a current image-portion within a current-field of a current-frame of said interlaced video data;
forming an interpolation of said image-portion in accordance with data within a remaining field of said current-frame; and
comparing said image-portion with said interpolation.
2. An intra-frame motion analyzer comprising:
means for receiving interlaced video data including a current image-portion within a current-field of a current-frame of said interlaced video data;
means for forming an interpolation of said image-portion in accordance with data within a remaining field of said current-frame; and
means for comparing said image-portion with said interpolation.
3. A method for determining inter-frame motion, comprising:
receiving interlaced video data including a current image-portion within a current-field of a current-frame of said interlaced video data; and
comparing said current image-portion with received video data including a corresponding image-portion of an immediately prior frame.
4. A method according to claim 3 , wherein
said image-portion is a current-pixel within a current scan-line, and
said corresponding image-portion is a corresponding pixel in a corresponding field of said immediately prior frame.
5. An inter-frame motion analyzer, comprising:
means for receiving interlaced video data including a current image-portion within a current-field of a current-frame of said interlaced video data; and
means for comparing said current image-portion with received video data including a corresponding image-portion of an immediately prior frame.