Free-Viewpoint Object Positioning for Intent-Aware Content Reproduction

The patent applies preliminary action by pre-defining reference viewpoints and their corresponding object position information before content reproduction. The system stores multiple sets of object position data corresponding to different reference viewpoints in advance, allowing the listener to move freely while the system selects and applies the appropriate pre-calculated position data based on the current listening position, thus conveying the content creator's intent without real-time complex calculations

The patent introduces reference viewpoints as an intermediary between the listener's position and the object arrangement. Instead of directly calculating object positions based on the listener's absolute position, the system uses reference viewpoints as intermediate reference frames. The object position information is stored relative to these reference viewpoints, which mediate the transformation from listener position to object arrangement, simplifying the overall system while enabling expressive content reproduction

The system performs preliminary calculations by pre-computing and storing object position information for multiple reference viewpoints before content reproduction. This allows the runtime system to simply select and interpolate between pre-computed data based on the listener's position, achieving high precision object positioning without requiring complex real-time calculations, thus reducing computational power requirements while maintaining accuracy

The patent segments the continuous listening space into discrete reference viewpoint regions. Instead of calculating object positions for every possible listener position, the system divides the space into regions associated with specific reference viewpoints and uses the corresponding pre-computed object position data for each region, reducing computational complexity while maintaining sufficient precision through interpolation between segments