Generalized scalability for video coder based on video objects

A video decoding system in which video objects are recognized from video data, wherein instances of a video object at given times are coded as video object planes (VOPs) and VOPs are assigned to one or more video object layers, the video decoding system comprising:

• a base layer decoder having an input for VOP data associated with a first video object layer of the video object;
• a processor coupled to an output of the base layer decoder;
• and
• an enhancement layer decoder, having a first input for VOP data associated with a second video object layer of the video object and a second input coupled to the processor and responsive to bidirectionally-predictive coded VOP (B-VOP) data including a ref_select_code included therein, the enhancement layer decoder decoding the B-VOP data with reference to forward and backward reference VOPs selected from the group of:
• forward:
• data of a VOP most recently decoded by the enhancement layer decoder;
• and
• data of a most recent VOP in display order decoded by the base layer decoder;
• and backward:
• data of a temporally coincident VOP decoded by the base layer decoder;
• data of a most recent VOP in display order decoded by the base layer decoder;
• and
• data of a next VOP in display order decoded by the base layer decoder.

A tangible computer-readable storage medium storing instructions for controlling the instructions causing the decoder to perform steps comprising:

• identifying data of a coded video object based on a start code indicative thereof;
• identifying data of a plurality of coded video object layers (VOLs) based on respective VOL start codes;
• identifying data of plurality of coded video object planes (VOPs) based on respective VOP start codes, each coded VOP being a member of at most one VOL;
• determining, from data of a first VOL:
• whether scalability had been applied to the first VOL coding;
• and
• if scalability had been applied, identifying from the coded data of the first VOL data a second VOL that is a reference layer to the first VOL;
• and
• decoding a VOP from the first VOL with reference to a VOP of the second video object layer.