Understanding human attention patterns and viewing strategies for UHD content is important for developing efficient data compression algorithms and accurate objective quality metrics. To help studying these properties and their dynamics when HD content is replaced with UHD content, we provide a publicly accessible dataset, called Ultra-Eye, which is composed of 41 4K UHD and HD images in TIFF format with corresponding eye tracking information. The eye tracking information includes the fixation points and fixation density maps measured during extensive subjective experiments. The dataset can be used for research purposes.
The dataset consists of 41 still images acquired using several cameras, including Sony DSC-RX100 II, Sony NEX-5N, FUJIFILM XF1, Olympus E-PL2, and RED SCARLET-X. Some of the images were obtained from Europeana internet portal1. The images cover a wide variety of scenes, including outdoor and indoor scenes, images of nature, people, animals, and historical scenes depicted in paintings.
Sample images from Ultra-Eye dataset
1Europeana: think culture, http://www.europeana.eu
Eye tracking experiments
The eye tracking experiments were conducted at the MMSPG test laboratory. The laboratory environments and the viewing conditions were set to fulfill recommendations ITU-R BT.500-13  and ITU-R BT.2022 , respectively. Please see the following table for the specific information on the experimental conditions.
|Participants||Number (M/F)||20 (13/7)|
|Age range (Ave.)||18-27 (23.8)|
|Screening||Snellen and Ishihara charts|
|Color temperature||6500 [K]|
|Viewing distance||1.12 [m]|
|Resolution||3840 × 2160 [pixels]|
|Angular resolution||60 [pixel/degree]|
|Eye tracker||Manufacturer||Smart Eye|
|Model||Smart Eye Pro 5.8|
|Mounting position||0.7 [m] from the display|
|Sampling frequency||60 [Hz]|
|Accuracy||< 0.5 [degree]|
|Calibration points||5 points on screen|
|Image presentation||Presentation order||Random|
|Presentation time||15 [s]|
|Grey-screen duration||2 [s]|
 ITU-R BT.500-13, “Methodology for the subjective assessment of the quality of television pictures,” International Telecommunication Union, January 2012.  ITU-R BT.2022, “General viewing conditions for subjective assessment of quality of SDTV and HDTV television pictures on flat panel displays,” International Telecommunication Union, August 2012.
Fixation density maps
The fixation density maps (FDMs) were computed by convolving the recorded gaze points with a Gaussian filter, and then normalizing the result to values between 0 and 1. Only gaze points corresponding to fixation points were used for computing a FDM. The standard deviation of the Gaussian filter used for computing the FDMs was set to 1 degree of visual angle, which corresponds to
If you use the Ultra-Eye dataset in your research, we kindly ask you to reference the following paper and URL link of this website:
URL link: http://mmspg.epfl.ch/ultra-eye
Permission is hereby granted, without written agreement and without license or royalty fees, to use, copy, modify, and distribute the data provided and its documentation for research purpose only. The data provided may not be commercially distributed. In no event shall the Ecole Polytechnique Fédérale de Lausanne (EPFL) be liable to any party for direct, indirect, special, incidental, or consequential damages arising out of the use of the data and its documentation. The Ecole Polytechnique Fédérale de Lausanne (EPFL) specifically disclaims any warranties. The data provided hereunder is on an “as is” basis and the Ecole Polytechnique Fédérale de Lausanne (EPFL) has no obligation to provide maintenance, support, updates, enhancements, or modifications.