3DVQM : 3d video quality monitor

• Autor(a) principal: Feitor, Bruno Tomé dos Santos
• Data de Publicação: 2013
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.8/3114
• topic: 3D video; Video quality assessment; Monitor’s software; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

This dissertation presents a research study and software implementation of an objective quality monitor for 3D video streams transmitted over networks with non-guaranteed packet delivery due to errors, congestion, excessive delay, etc. A review of Video Quality Assessment (VQA) models available in the literature is first presented, addressing 2D and 3D video quality models that were selected as relevant for this research work. A packet-layer VQA model is proposed based on header information from three different packet-layer levels: Network Abstraction Layer (NAL), Packetised Elementary Streams (PES) and MPEG2 - Transport Stream (TS). Transmission errors leading to undecodable TS packets are assumed to result in a whole frame loss. The proposed method estimates the size of the lost frames, which is used as a model parameter to predict their objective quality, measured as the Structural Similarity Index Metric (SSIM). In order to materialise the proposed VQA model, a software application was developed that allows monitoring a corrupted 3D video stream quality. To make the monitoring process as user friendly as possible, a Guide User Interface (GUI) was developed. With this feature the user can interact with the application by controlling the input parameters and customizing the results on the output display. The results show that SSIM of isolated missing stereoscopic frames in 3D coded video can be predicted with Root Mean Square Error (RMSE) accuracy of about 0.1 and Pearson correlation coefficient of 0.8, taking the SSIM of uncorrupted frames as reference. It is concluded that the proposed model is capable of estimating the SSIM quite accurately using only the estimated sizes of single lost frames.