Abstract

Efficient representation and coding of fine-granular motion information is one of the key research areas for exploiting inter-frame correlation in video coding. Representative techniques towards this direction are affine motion compensation (AMC), decoder-side motion vector refinement (DMVR), and subblock-based temporal motion vector prediction (SbTMVP). Fine-granular motion information is derived at subblock level for all the three coding tools. In addition, the obtained inter prediction can be further refined by two optical flow-based coding tools, the bi-directional optical flow (BDOF) for bi-directional inter prediction and the prediction refinement with optical flow (PROF) exclusively used in combination with AMC. The aforementioned five coding tools have been extensively studied and finally adopted in the Versatile Video Coding (VVC) standard. This paper presents technical details of each tool and highlights the design elements with the consideration of typical hardware implementations. Following the common test conditions defined by Joint Video Experts Team (JVET) for the development of VVC, 5.7% bitrate reduction on average is achieved by the five tools. For test sequences characterized by large and complex motion, up to 13.4% bitrate reduction is observed. Additionally, visual quality improvement is demonstrated and analyzed.

Highlights

  • INTRODUCTIONV IDEO coding standards play an increasingly important role in diversified video applications and services ranging from the conventional television broadcasting, internet

  • V IDEO coding standards play an increasingly important role in diversified video applications and services ranging from the conventional television broadcasting, internetManuscript received August 22, 2020; revised March 25, 2021 and June 17, 2021; accepted July 17, 2021

  • The Versatile Video Coding (VVC) standard [1] was developed by the Joint Video Experts Team (JVET) to face the technical challenge of higher compression ratio for the exponentially increasing traffic of video content and to meet the requirements from emerging markets including the products and services with featured content such as high dynamic range (HDR), augmented reality (AR) and virtual reality (VR), etc

Read more

Summary

INTRODUCTION

V IDEO coding standards play an increasingly important role in diversified video applications and services ranging from the conventional television broadcasting, internet. In contrast to the sample-wise processing in traditional affine schemes, the introduction of subblocks makes a good trade-off of coding efficiency and computational complexity, and reduces the implementation cost by reusing existing hardware or software modules for block-based motion compensation Another way of reducing the overhead of transmitting motion parameters is to derive motion parameters at the decoder side [24]–[28]. The Subblock-based TMVP (SbTVMP) was initially proposed in the exploration stage for the new video coding standard after HEVC [32] and was adopted in JEM It allows inheriting the motion information at subblock level from the collocated reference picture.

SUBBLOCK-BASED MOTION DERIVATION
Affine Motion Compensation
Subblock-Based Temporal Motion Vector Prediction
Decoder-Side Motion Vector Refinement
Bi-Directional Optical Flow for CU-Based Inter Prediction
Affine Prediction Refinement With Optical Flow
IMPLEMENTATION CONSIDERATIONS
Bi-Directional Optical Flow
EXPERIMENTAL RESULTS AND ANALYSIS
Statistics Analysis
Visual Quality Impact
CONCLUSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call