Efficient modeling of video-to-image conversion for QoS management in networked video service

Modality conversion (also called transmoding) currently emerges as an important issue in universal multimedia access. Intuitively, when content scaling cannot provide the acceptable QoS, modality conversion may be the good choice to maintain the quality. For this purpose, we need to answer a QoS-related question "at which resource constraint should the current modality be converted?" In our previous works, we have proposed the overlapped content value model that represents the relationship between quality and resource for different modalities, thus helping to find the conversion points between modalities. In this paper, we propose an efficient method to build the OCV model in the case of video-to-image conversion.