Adaptive bit rate allocation

摘要

A method for determining quantization numbers for each macro block in one video segment having a prescribed capacity is disclosed. The quantization numbers determine how much data will be preserved for that macro block. The method begins by determining a level of complexity for each macro block. Next, initial quantization numbers are chosen for the macro blocks by choosing the largest values possible without exceeding the prescribed capacity of the video segment. Final quantization numbers are selected based on respective ones of the initial quantization numbers proportioned according to the level of complexity for that macro block. The final quantization numbers may be increased or decreased so that the capacity of the video segment is maximized but not exceeded.