Semantics-based motion estimation for multi-view video coding

A computer readable medium that provides instructions, which when executed on a processor cause the processor to perform a method comprising:

• identifying one or more pixels in a first frame of a multi-view video sequence;
• constraining a search range associated with a second frame of the multi-view video sequence based upon an indication of a desired correlation between efficient coding and semantic accuracy, the semantic accuracy relying on use of geometric configurations of cameras capturing the multi-view video sequence;
• and
• searching the second frame within the constrained search range for a match of the one or more pixels identified in the first frame.

A computerized system comprising:

• a memory;
• and
• at least one processor coupled to the memory, the at least one processor executing a set of instructions which cause the at least one processor to
• identify one or more pixels in a first frame of a multi-view video sequence,
• constrain a search range associated with a second frame of the multi-view video sequence based upon an indication of a desired correlation between efficient coding and semantic accuracy, the semantic accuracy relying on use of geometric configurations of cameras capturing the multi-view video sequence, and
• search the second frame within the constrained search range for a match of the one or more pixels identified in the first frame.

A motion estimation apparatus comprising:

• a block identifier to identify one or more pixels in a first frame of a multi-view video sequence;
• a search range determinator to constrain a search range associated with a second frame of the multi-view video sequence based upon an indication of a desired correlation between efficient coding and semantic accuracy, the semantic accuracy relying on use of geometric configurations of cameras capturing the multi-view video sequence;
• and
• a searcher to search the second image within the constrained search range for a match of the one or more pixels identified in the first frame.