VideoSet

A new methodology to measure coded image/video quality using the just-noticeable-difference (JND) idea was proposed in [1].  Several small JND-based image/video quality datasets were released by the Media Communications Lab at the University of Southern California in [2, 3]. In this work, we present an effort to build a large-scale JND-based coded video quality dataset. The dataset consists of 220 5-second sequences in four resolutions (i.e., 1920x1080, 1280x720, 960x540, and 640x 360).  For each of the 880 video clips, we encode it using the H.264 codec with QP=1, ..., 51 and measure the first three JND points with 30+ subjects.  The dataset is called the ``VideoSet", which is an acronym for ``Video Subject Evaluation Test (SET)". This work describes the subjective test procedure, detection and removal of outlying measured data, and the properties of collected JND data. Finally, the significance and implications of the VideoSet to future video coding research and standardization efforts are pointed out. All source/coded video clips as well as measured JND data included in the VideoSet are available to the public here.

 [1] Joe Yuchieh Lin, et al., Experimental design and analysis of JND test on coded image/video. Proc. SPIE, Applications of Digital Image Processing XXXVIII, September 22, 2015. [2] L. Jin, et al., Statistical study on perceived JPEG image quality via MCL- JCI dataset construction and analysis, in: IS&T/SPIE Electronic Imaging, International Society for Optics and Photonics, 2016. [3] H. Wang, et al., MCL-JCV: A JND-based H.264/AVC video quality assessment dataset, in: 2016 IEEE International Conference on Image Processing (ICIP), 2016, pp. 1509–1513.

Update:

• As requested by some users, we prepared a copy of videoSet hosted on BaiduYun. Please try both servers, i.e. AWS in this page and BaiduYun and find the one that works better for you. Date: 02/14/2017.
• A typo fixed. The second header in 960x540_1st.csv, changed from anchor to jnd. Date: 03/01/2017
• Citation was updated in README_V2. The dataset paper was accepted by JVCIR and PDF attached. Date: 05/09/2017.