In this paper, we propose a fast mode decision framework and a fast motion estimation algorithm for H.264 to High Efficiency Video Coding (HEVC) transcoding. The fast mode decision framework employs a post-order (bottom-up) traversal of the coding tree unit (CTU) quadtree. Based on this traversal and H.264 information, several strategies are proposed to reduce HEVC modes to be tested and a ratedistortion (RD) cost prediction model is used to terminate the processing of a tested mode early. The proposed fast motion estimation algorithm selects the best candidate from a list of H.264 motion vectors (MVs) and previously encoded HEVC MVs. Compared to a full re-encoding, experimental results show that the proposed solution achieves speed-ups of up to 12.75x, for an average BD-Rate of 3.28%.

Abstract-This paper describes a novel multi-frame and multislice parallel video encoding approach with simultaneous encoding of predicted frames. The approach, when applied to H.264 encoding, leads to speedups comparable to those obtained by state-of-the-art approaches, but without the disadvantage of requiring bidirectional frames. The new approach uses a number of slices equal or greater than the number of cores used and supports three motion estimation modes. Their combination leads to various tradeoffs between speedup and visual quality loss. For an H.264 baseline profile encoder based on Intel IPP code samples running on a two quad core Xeon system (8 cores in total), our experiments show an average speedup of 7.20x, with an average quality loss of 0.22 dB (compared to a nonparallelized version) for the most efficiency motion estimation mode, and an average speedup of 7.95x, with a quality loss of 1.85 dB for the faster motion estimation mode

In this paper, we propose a fast H.264-to-HEVC transcoder composed of a motion propagation algorithm and a fast mode decision framework. The motion propagation algorithm creates a motion vector candidate list at the coding tree unit (CTU) level and, thereafter, selects the best candidate at the prediction unit level. This method eliminates computational redundancy by pre-computing the prediction error of each candidate at the CTU level and reusing the information for various partition sizes. The fast mode decision framework is based on a post-order traversal of the CTU, and includes several mode reduction techniques. In particular, the framework permits the early termination of the rate distortion cost computation, a highly complex task, when a mode is unpromising. Moreover, a novel method exploits the data created by the motion propagation algorithm to determine whether a coding unit (CU) must be split. This allows the pruning of unpromising sub-partitions. Compared to a cascaded pixel-domain transcoding approach, the experimental results show that the proposed solution using one reference frame is on average 8.5 times faster, for an average BD-Rate of 2.63%. For a configuration with 4 reference frames, the average speed-up is 11.77x and the average BD-Rate is 3.82%.