Effect of Codec Performance on Video QoE for videos encoded with Xvid, H.264 and WebM/VP8

Abstract: In recent years, there has been a significant growth in multimedia services such as mobile video streaming, Video-on-Demand and video conferencing. This has led to the development of various video coding techniques, aiming to deliver high quality video while using available bandwidth efficiently. This upsurge in the usage of video applications has also resulted in making endusers more quality-conscious. In order to meet the users’ expectations, the Quality of Experience (QoE) studies has gained utmost importance from both researchers and service providers. This thesis aims to compare the performance of H.264/AVC, Xvid and WebM/VP8 video codecs in wired and wireless networks. The codec performance is evaluated for different packet loss and delay variation values. The evaluation of codec performance is done using both subjective and objective assessment methods. In subjective assessment method, the evaluation of video codec performance is done using ITU-T recommended Absolute Category Rating (ACR) method. Using this method the perceptual video quality ratings are taken from the users, which are then averaged to obtain Mean Opinion Score. These obtained scores are used to analyze the performance of encoded videos with respect to users’ perception. In addition to subjective assessment method, the quality of encoded video is also measured using objective assessment method. The objective metric SSIM (Structural Similarity) is used to evaluate the performance of encoded videos. Based on the results, it was found that for lower packet loss and delay variation values H.264 showed better results when compared to Xvid and WebM/VP8 whereas, WebM/VP8 outperformed Xvid and H.264 for higher packet loss and delay variation values. On the whole, H.264 and WebM/VP8 performed better than Xvid. It was also found that all three video codecs performed better in wired network when compared to the wireless network.