Waterloo IVC 3D Image Quality Database
Phase I and Phase II
Descriptions:
- The Waterloo IVC 3D Image Quality database Phase I is created from 6 pristine stereoscopic image pairs (and thus their corresponding single-view images), all collected from the Middlebury Stereo 2005 Datasets. Each single-view image was altered by three types of distortions:additive white Gaussian noise contamination, Gaussian blur, and JPEG compression. Each distortion type had four distortion levels, where the distortion control parameters were decided to ensure a good perceptual separation. The single-view images were employed to generate distorted stereopairs, either symmetrically or asymmetrically. Altogether, there are totally 78 single-view images and 330 stereoscopic images.
- The Waterloo IVC 3D Image Quality database Phase II with more diverse image content is created from 10 pristine stereoscopic image pairs (and thus their corresponding single-view images). All images were collected from previous subjective 3D quality studies [1] ,[2] and [3]. Each single-view image was altered by the same three types of distortions and each distortion type had the same four distortion levels. There are totally 130 single-view images and 460 stereoscopic images.
Publications:
- Jiheng Wang, Abdul Rehman, Kai Zeng, Shiqi Wang and Zhou Wang, "Quality prediction of asymmetrically distorted stereoscopic 3D images", IEEE Transactions on Image Processing, vol. 24, no. 11, pp. 3400-3414, Nov. 2015. [pdf]
- Jiheng Wang, Kai Zeng and Zhou Wang, "Quality prediction of asymmetrically distorted stereoscopic images from single views", IEEE International Conference on Multimedia and Expo (oral presentation, 13.7% acceptance rate), Chengdu, Sichuan, China, July 14-18, 2014. [pdf][slides][poster]
- Jiheng Wang and Zhou Wang, "Perceptual quality of asymmetrically distorted stereoscopic images: the role of image distortion types", International Workshop on Video Processing and Quality Metrics for Consumer Electronics (invited paper) , Chandler, AZ, Jan. 29-31, 2014. [pdf][slides]
- Jiheng Wang, Qingbo Wu, Abdul Rehman, Shiqi Wang and Zhou Wang, "Blind quality prediction of stereoscopic 3D images", IS&T Electronic Imaging: Human Vision and Electronic Imaging, Burlingame, CA, Jan. 2017. [pdf][slides]
References:
[1] M. Urvoy, M. Barkowsky, R. Cousseau, Y. Koudota, V. Ricorde, P. Le Callet, J. Gutiérrez, and N. García, “NAMA3DS1-COSPAD1: Subjective video quality assessment database on coding conditions introducing freely available high quality 3D stereoscopic sequences”, International Workshop on Quality of Multimedia Experience, Yarra Valley, VIC, Australia, July 5-7, 2012.
[2] M. H. Pinson, “The consumer digital video library [best of the web]”, IEEE Signal Processing Magazine, vol. 30, no. 4, pp. 172–174, July 2013.
[3] ISO/IEC JTC1/SC29/WG11, “Undo Dancer 3DV sequence for purposes of 3DV standardization,” Document M20028, ISO, Geneva, Switzerland, 2011.
Download Now: Phase I [1.43 GB], Phase II [2.90 GB] and Matlab Code [62.4KB]
Password: uwivc3D2015