Reducing and correcting motion estimation artifacts during video frame rate conversion

A method comprising:

• projecting motion vectors describing a transformation from a previous video frame to a future video frame onto a plane between the previous video frame and the future video frame, the previous video frame being in a temporal past compared to the future video frame;
• statistically analyzing the projected motion vectors for a number of cover regions, a number of uncover regions, and a number of conflicting regions on the plane, the cover region indicating that corresponding pixels on a region of the plane are solely located in the previous video frame, the uncover region indicating that corresponding pixels on the region of the plane are solely located in the future video frame, and the conflicting region simultaneously indicating a cover region and an uncover region;
• analyzing a dissimilarity between a trial video frame and both the previous video frame and the future video frame, the trial video frame being generated between the previous video frame and the future video frame based on a frame rate conversion ratio derived from a source frame rate and a desired frame rate;
• correcting a location of the plane and a blending factor for a pixel on the plane based on the statistical analysis and the dissimilarity analysis through obtaining a figure of demerit for each N×N block of pixels at the plane based on a higher of a result of the potential artifact detection and a result of the dissimilarity analysis to enable concealment of the corresponding N×N block of pixels, N being a positive even number;
• and
• generating, between the previous video frame and the future video frame, an interpolated video frame with the corrected blending factor at the corrected location of the plane.

A system comprising:

• a display unit configured to display a video at a first frame rate;
• and
• a multimedia processor configured to:
• project motion vectors describing a transformation from a previous video frame to a future video frame onto a plane between the previous video frame and the future video frame, the previous video frame being in a temporal past compared to the future video frame, and both the previous video frame and the future video frame being frames of a source video configured to be rendered on the display unit,
• detect potential artifacts at the plane based on an intersection of a cover region and an uncover region on the plane, the cover region indicating that corresponding pixels on a region of the plane are solely located in the previous video frame, the uncover region indicating that corresponding pixels on the region of the plane are solely located in the future video frame, and the intersection of the cover region and the uncover region indicating a plurality of conflicting motion vectors at a region of intersection on the plane,
• analyze a dissimilarity between a trial video frame and both the previous video frame and the future video frame, the trial video frame being generated between the previous video frame and the future video frame based on a frame rate conversion ratio derived from a frame rate of the source video and the first frame rate,
• correct a location of the plane and a blending factor for a pixel on the plane based on the potential artifact detection and the dissimilarity analysis through obtaining a figure of demerit for each N×N block of pixels at the plane, the figure of merit being based on a higher of a result of the potential artifact detection and a result of the dissimilarity analysis to enable concealment of the corresponding N×N block of pixels, and N being a positive even number, and
• generate, between the previous video frame and the future video frame, an interpolated video frame with the corrected blending factor at the corrected location of the plane to be rendered along with the previous video frame and the future video frame on the display unit at the first frame rate.