Direct and Semi-Direct Frame Partitioning Mapping for Computation and Energy-Aware HEVC Transrating

Abstract

The most common operation to maintain various versions of the same video sequence under different bitrates in live and streaming video applications is video transrating. The typical transrating operation is comprised of a decoding and a complete encoding step in sequence, which results in long processing times and high energy consumption. This work proposes two early-termination schemes based on frame partitioning inheritance between decoding and encoding to accelerate the complex quadtree structure decisions performed during HEVC transrating. The proposed Direct Mapping scheme achieved time saving results of 48.5%, with a compression efficiency loss of 0.818%, on average. The Semi-Direct Mapping reaches 40.2% in time savings at the cost of an average compression efficiency loss of 0.724% in comparison to the original transcoder. Energy consumption was reduced in 50.4%, on average, considering the Direct Mapping approach.