A Decoding-Complexity and Rate-Controlled Video-Coding Algorithm for HEVC

• Published in: Future internet 2020-07, Vol.12 (7), p.120
• Main Authors: Mallikarachchi, Thanuja, Talagala, Dumidu, Kodikara Arachchi, Hemantha, Hewage, Chaminda, Fernando, Anil
• Format: Article
• Language: English
• Subjects: decoding-complexity control; decoding-complexity–rate–distortion; decoding-energy; energy consumption control; HEVC; rate control
• ISSN: 1999-5903; 1999-5903
• DOI: 10.3390/fi12070120

Video playback on mobile consumer electronic (CE) devices is plagued by fluctuations in the network bandwidth and by limitations in processing and energy availability at the individual devices. Seen as a potential solution, the state-of-the-art adaptive streaming mechanisms address the first aspect, yet the efficient control of the decoding-complexity and the energy use when decoding the video remain unaddressed. The quality of experience (QoE) of the end-users’ experiences, however, depends on the capability to adapt the bit streams to both these constraints (i.e., network bandwidth and device’s energy availability). As a solution, this paper proposes an encoding framework that is capable of generating video bit streams with arbitrary bit rates and decoding-complexity levels using a decoding-complexity–rate–distortion model. The proposed algorithm allocates rate and decoding-complexity levels across frames and coding tree units (CTUs) and adaptively derives the CTU-level coding parameters to achieve their imposed targets with minimal distortion. The experimental results reveal that the proposed algorithm can achieve the target bit rate and the decoding-complexity with 0.4% and 1.78% average errors, respectively, for multiple bit rate and decoding-complexity levels. The proposed algorithm also demonstrates a stable frame-wise rate and decoding-complexity control capability when achieving a decoding-complexity reduction of 10.11 (%/dB). The resultant decoding-complexity reduction translates into an overall energy-consumption reduction of up to 10.52 (%/dB) for a 1 dB peak signal-to-noise ratio (PSNR) quality loss compared to the HM 16.0 encoded bit streams.