This dataset contains 1 audio, two video and three audiovisual databases. They were produced for training and validating the standardized models ITU-T P.1201.2 and P.1202.2. Both models are bitstream-based no-reference models targeting IPTV services and high-resolution video sequences (Standard and High Definition). They cover compression and packet-loss impairments. Each database contains about 240 sequences. For video, the H.264 codec was used, while the audio was encoded with four different codecs (AAC-LC, HE-AAC, MP2, AC-3). Both uniform (random) and bursty losses were addressed for different types of packet-loss concealment. Databases include per-subject quality scores as well as per-frame (video and audio) and per sequence bitstream- and packet loss-related information.
The dataset is available at the following URL: ftp://ftp.ivc.polytech.univ-nantes.fr/VQEG/JEG/HYBRID/database_TLABS_P1201_IPTVH264HD_Audiovisual/
References and Citation
Please cite the following paper in your reference if you use this dataset for your work [GLF2016].
GLF2016: M.-N. Garcia, P. List, B. Feiten, U. Wüstenhagen and A. Raake (2016). Audio-Video Databases for H.264-bitstream-based Quality Assessment of IPTV Services, 8th International Workshop on Quality of Multimedia Experience (QoMEX).