Deep Learning-Based Luma and Chroma Fractional Interpolation in Video Coding

• Published in: IEEE access 2019, Vol.7, p.112535-112543
• Main Authors: Pham, Chi Do-Kim, Zhou, Jinjia
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Coding standards; Convolution neural network (CNN); Deep learning; Encoding; fractional interpolation; Image resolution; Interpolation; motion compensated prediction; Redundancy; Signal resolution; Training; Video coding
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2019.2935378

Motion compensated prediction is one of the essential methods to reduce temporal redundancy in inter coding. The target of motion compensated prediction is to predict the current frame from the list of reference frames. Recent video coding standards commonly use interpolation filters to obtain sub-pixel for the best matching block located in the fractional position of the reference frame. However, the fixed filters are not flexible to adapt to the variety of natural video contents. Inspired by the success of Convolutional Neural Network (CNN) in super-resolution, we propose CNN-based fractional interpolation for Luminance (Luma) and Chrominance (Chroma) components in motion compensated prediction to improve the coding efficiency. Moreover, two syntax elements indicate interpolation methods for the Luminance and Chrominance components, have been added to bin-string and encoded by CABAC using regular mode. As a result, our proposal gains 2.9%, 0.3%, 0.6% Y, U, V BD-rate reduction, respectively, under low delay P configuration.