A Convolutional Neural Network Approach for Objective Video Quality Assessment
IEEE Transactions on Neural Networks
User-oriented QoS in packet video delivery
IEEE Network: The Magazine of Global Internetworking
| Hi-index | 0.00 |
We propose a no reference hybrid video-quality-estimation model for estimating video quality by using quality features derived from received packet headers and video signals. Our model is useful as a quality monitoring tool for estimating the video quality during use of an Internet protocol television service. It takes into account video quality dependence on video content and can estimate video quality per content, which our previously developed packet-layer model cannot do. We conducted subjective quality assessments to develop the model and validated its quality-estimation accuracy. The quality-estimation results showed that the Pearson-correlation coefficients were larger than 0.9 and the quality-estimation errors were equivalent to the statistical uncertainty of subjective quality.