Probabilistic Approach Versus Machine Learning for One-Shot Quad-Tree Prediction in an Intra HEVC Encoder

Abstract

Evolutions of the Internet of Things (IoT) in the next years are likely to boost mobile video demand to an unprecedented level. A large number of battery-powered systems will integrate an HEVC video codec, implementing the latest encoding MPEG standard, and these systems will need to be energy efficient. Constraining the energy consumption of HEVC encoders is a challenging task, especially for embedded applications based on software encoders. The most efficient approach to reduce the energy consumption of an HEVC encoder consists in optimizing the quad-tree block partitioning of the image and trade-off compression efficiency and energy consumption by efficiently choosing the near-optimal pixel block sizes. For the purpose of reducing the energy consumption of a real-time HEVC Intra encoder, this paper proposes and compares two methods that predict the quad-tree partitioning in "one-shot", i.e. without iterating. These methods drastically limit the computational cost of the recursive Rate-Distortion Optimization (RDO) process. The first proposed method uses a Probabilistic approach whereas the second method is based on Machine Learning approach. Experimental results show that both methods are capable of reducing the energy consumption of an embedded HEVC encoder of 58% for a bit rate increase of respectively 3.93% and 3.6%.

QoE-Based Performance Comparison of AVC, HEVC, and VP9 on Mobile Devices with Additional Influencing Factors

“…All four videos are different from each other in terms of spatial characteristics. The spatial and temporal details of these videos can be found in these articles [48][49][50]. AVC and HEVC videos were encoded using libx264 and libx265 video coding libraries, respectively, with Matroska [51] as the container, whereas the VP9 videos were encoded using the libvpx-vp9 library and WEBM [28] as the container.…”

Efficient VVC Intra Prediction Based on Deep Feature Fusion and Probability Estimation

“…However, compared with the computational complexity of the encoder, the overhead of ML-based algorithm is negligible [19]. In [20], Mercat et al demonstrated the superiority of MLbased method to conventional probability-based method in the scenario of fast video coding. As a popular tool in machine learning, the random forest was demonstrated to effectively predict the intra angular prediction mode by Ryu et al [21].…”