Abstract | ||
---|---|---|
In this paper, a traffic reduction scheme for video-based call and chat applications that uses deep neural network (DNN) based super resolution is proposed. Specifically, a sender transmits low-quality, low-resolution video frames containing face information in order to reduce the amount of video traffic. The receiver uses DNN-based super resolution to reconstruct high-quality, high-resolution video frames from the low-quality video frames. The proposed scheme makes two contributions. First, face features are adopted for parameter optimization of DNN-based super resolution for high-quality image reconstruction, and second, the scheme includes a newly designed loss function that considers face features that allow high-quality face-containing video frames to be reconstructed at the receiver. According to our evaluation results using real video frames of video calls, the proposed scheme reduces the amount of video traffic by more than 90% as compared with conventional schemes that implement the standard video encoder. In this case, the proposed scheme achieves a reconstructed image quality up to 0.85 in terms of structural similarity (SSIM). |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/ICC.2019.8761766 | IEEE International Conference on Communications |
Keywords | Field | DocType |
Deep Neural Network,Video-based Call and Chat,Super-Resolution | Iterative reconstruction,Computer science,Image quality,Communication source,Real-time computing,Encoder,Traffic reduction,Artificial neural network,Superresolution | Conference |
ISSN | Citations | PageRank |
1550-3607 | 0 | 0.34 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Shota Watanabe | 1 | 0 | 0.68 |
Takuya Fujihashi | 2 | 34 | 11.65 |
Shunsuke Saruwatari | 3 | 64 | 14.40 |
Takashi Watanabe | 4 | 165 | 22.64 |