Non-intrusive Packet-Layer Model for Monitoring Video Quality of IPTV Services

この論文をさがす

説明

Developing a non-intrusive packet-layer model is required to passively monitor the quality of experience (QoE) during service. We propose a packet-layer model that can be used to estimate the video quality of IPTV using quality parameters derived from transmitted packet headers. The computational load of the model is lighter than that of the model that takes video signals and/or video-related bitstream information such as motion vectors as input. This model is applicable even if the transmitted bitstream information is encrypted because it uses transmitted packet headers rather than bitstream information. For developing the model, we conducted three extensive subjective quality assessments for different encoders and decoders (codecs), and video content. Then, we modeled the subjective video quality assessment characteristics based on objective features affected by coding and packet loss. Finally, we verified the model's validity by applying our model to unknown data sets different from training data sets used above.

収録刊行物

被引用文献 (5)*注記

もっと見る

参考文献 (30)*注記

もっと見る

詳細情報 詳細情報について

問題の指摘

ページトップへ