Video Block Multi-Hypothesis Prediction for Coding Efficiency

The patent segments the prediction process into multiple hypotheses, where each hypothesis represents a different motion candidate from the non-regular merge list. This segmentation allows the system to evaluate multiple motion possibilities independently and combine them to achieve better coding efficiency while maintaining manageable complexity through structured processing.

The patent introduces an additional dimension to the motion prediction by incorporating multiple hypotheses beyond the conventional single motion candidate approach. This multi-hypothesis framework extends the prediction space, allowing the encoder to explore multiple motion scenarios simultaneously, thereby improving coding efficiency without significantly increasing implementation complexity.

The patent performs preliminary action by pre-defining the non-regular merge list and motion candidates before the actual prediction process. This preparation work is done once and reused across multiple predictions, reducing the repeated computational overhead and allowing the multi-hypothesis approach to achieve better compression performance with controlled computational power consumption.

The patent applies partial action by selectively applying the multi-hypothesis prediction process only to specific video blocks where it provides significant benefit, rather than uniformly applying it to all blocks. This selective approach improves overall compression performance while limiting the increase in computational power to only where necessary.