Defining a search range for motion estimation for each scenario frame set

A video motion estimation apparatus, comprising:

• processing circuitry configured to:
• obtain a plurality of image frames in a video, and perform scenario classification processing on the plurality of image frames by determining a background image region of each of the plurality of image frames and an image similarity between each of the background image regions to obtain a plurality of image frame sets, each image frame set being associated with and classified based on a corresponding scenario and comprising plural image frames;
• extract a contour feature and a color feature of a foreground object of each of the plurality of image frame sets;
• determine a search range corresponding to each of the plurality of image frame sets based on the contour feature and the color feature of the foreground object of the image frames in the respective image frame set;
• determine a starting search point in each predicted frame in each of the plurality of image frame sets, wherein the respective image frame set comprises at least one predicted frame and at least one reference frame;
• and
• for each of the plurality of image frame sets, perform motion estimation processing in a search region corresponding to the search range of each predicted frame in the respective image frame set based on the starting search point of the respective predicted frame, a reference block in the at least one reference frame of the respective image frame set, and the color feature of the foreground object, to obtain a motion vector corresponding to the reference block.

A non-transitory computer-readable storage medium storing computer-readable instructions thereon, which, when executed by a processor, cause the processor to perform a video motion estimation method comprising:

• obtaining a plurality of image frames in a video, and performing scenario classification processing on the plurality of image frames by determining a background image region of each of the plurality of image frames and an image similarity between each of the background image regions to obtain a plurality of image frame sets, each image frame set being associated with and classified based on a corresponding scenario and comprising plural image frames;
• extracting a contour feature and a color feature of a foreground object of each image frame in each of the plurality of image frame sets;
• determining a search range corresponding to each of the plurality of image frame sets based on the contour feature and the color feature of the foreground object of the image frames in the respective image frame set;
• determining a starting search point in each predicted frame in each of the plurality of image frame sets, wherein the respective image frame set comprises at least one predicted frame and at least one reference frame;
• and
• for each of the plurality of image frame sets, performing motion estimation processing in a search region corresponding to the search range of each predicted frame in the respective image frame set based on the starting search point of the respective predicted frame, a reference block in the at least one reference frame of the respective image frame set, and the color feature of the foreground object, to obtain a motion vector corresponding to the reference block.