Intra-mode Selection Without Reconstructed Data

The patent applies preliminary action by performing direction estimation for the next block concurrently with encoding the current block. The direction for the next block is determined in advance based on current block data and prediction models, so that when the next block is processed, the direction is already ready, eliminating the need for complex post-processing and reducing overall computational load and power consumption.

The patent implements continuity of useful action through pipelined processing where direction estimation and block encoding operations overlap in time. The direction estimation unit continuously operates on current blocks while the encoding unit processes previously estimated directions, creating a continuous workflow that maximizes processor utilization and reduces idle power consumption without sacrificing encoding accuracy.

The patent applies segmentation by dividing the video encoding process into distinct functional units: a direction estimation unit that analyzes current blocks and predicts directions for next blocks, and an encoding unit that performs actual encoding using these pre-estimated directions. This segmentation allows the two operations to proceed concurrently on different data, increasing processing speed while keeping each unit's computational complexity manageable through specialized algorithms.

The direction estimation unit performs preliminary computation by determining directions for next blocks before the encoding unit needs to process them. This preliminary action shifts the computational workload to a different time slot, allowing the encoding unit to operate faster without waiting for direction calculations, thus increasing overall processing speed while distributing computational complexity across time rather than requiring simultaneous complex operations.

The patent applies preliminary action by estimating directions for next blocks based on current block data and prediction models before actual encoding occurs. This preliminary direction estimation eliminates the need to wait for reconstructed data from previous blocks, achieving encoding accuracy through predictive models rather than relying on completed reconstruction processes, thus reducing processing time while maintaining precision.

The patent uses copying by creating prediction models that replicate the characteristics of previously encoded blocks without actually processing the full reconstructed data. Instead of using the complete reconstructed data from previous blocks, the system copies essential directional information and statistical properties into prediction models, allowing fast intra-mode selection based on these simplified copies rather than full data reconstruction, thereby reducing processing time while maintaining encoding accuracy.