2021 IEEE International Conference on Image Processing
19-22 September 2021 • Anchorage, Alaska, USA
Imaging Without Borders
| Paper ID | COM-2.6 | ||
| Paper Title | CONVOLUTIONAL NEURAL NETWORK BASED IN-LOOP FILTER FOR VVC INTRA CODING | ||
| Authors | Yue Li, Li Zhang, Kai Zhang, Bytedance Inc., United States | ||
| Session | COM-2: Learning-based Image and Video Coding | ||
| Location | Area H | ||
| Session Time: | Wednesday, 22 September, 14:30 - 16:00 | ||
| Presentation Time: | Wednesday, 22 September, 14:30 - 16:00 | ||
| Presentation | Poster | ||
| Topic | Image and Video Communications: Lossy coding of images & video | ||
| IEEE Xplore Open Preview | Click here to view in IEEE Xplore | ||
| Abstract | In the emerging Versatile Video Coding (VVC) standard, there are three in-loop filters known as deblocking, sam- ple adaptive offset (SAO), and adaptive loop filter (ALF) for suppressing compression artifacts as well as reduc- ing distortion. However, those handcrafted filters are insufficient to deal with the complicated compression ar- tifacts. Deep learning-based filtering has demonstrated overwhelming successes in the field of image restoration. In this paper, we propose a convolutional neural network- based filter for enhancing the quality of VVC intra coded frames. The proposed filter takes auxiliary information including partitioning and prediction information as in- put. For chroma, auxiliary information further includes luma samples. Regarding the training, we find that data augmentation and loss function selection are not triv- ial. We conduct extensive experiments to evaluate the effectiveness of each design. When testing on top of VTM-10.0 under all intra configuration, the proposed fil- ter achieves state-of-the-art performance, i.e. on average 7.57%, 13.18%, and 12.50% BD-rate reductions for Y, Cb, and Cr, respectively. The proposed filter ranks first among neural network-based in-loop filters at the 20th meeting of the Joint Video Exploration Team (JVET). | ||