Understanding visual attention in 3DTV is essential for many applications, e.g., capture, coding, visual confort enhancement, 2D-to-3D conversion, retargeting, and subtitling. Therefore, public datasets of 3D content with associated ground truth eye tracking data are needed. To overcome the lack of publicly available 3D video eye tracking datasets, we created the EyeC3D dataset. Eight stereoscopic video sequences were used in the eye tracking experiments. For each video, eye movement data was recorded via a set of subjective experiments. From the eye movement data, the fixation density maps (FDMs) were computed for each frame of the stereoscopic video sequences.
Stereoscopic video sequences
Eight stereoscopic video sequences were used in the eye tracking experiments. Five sequences (Boxers, Hall, Lab, News report, and Phone call) were obtained from the NAMA3DS1 database . Two sequences (Musicians and Poker) were obtained from the European FP7 Research Project MUSCADE . Sequence Poznan Hall2 was obtained from the Poznan multiview video database .
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 recommendation ITU-R BT.2021 . Please see the following table for the specific information on the experimental conditions.
|Participants||Number (M/F)||21 (16/5)|
|Age range (Ave.)||18 − 31 (21.8)|
|Screening||Snellen chart, Ishihara chart, and Randot test|
|Color temperature||6500 [K]|
|Viewing distance||1.8 [m]|
|Resolution||1920 × 1080 [pixels]|
|Angular resolution||60 [pixel/degree]|
|Display calibration||Probe||X-Rite i1Display Pro|
|Profile||D65 white point, 120 [cd/m2] brightness, minimum black level|
|Eye tracker||Manufacturer||Smart Eye|
|Model||Smart Eye Pro 5.8|
|Mounting position||1.28 [m] from the display|
|Sampling frequency||60 [Hz]|
|Accuracy||< 0.5 [degree]|
|Calibration points||4 points on screen|
|Video presentation||Presentation order||Random|
|Presentation time||8 − 10 [s]|
|Grey-screen duration||2 [s]|
Fixation density maps
All detected saccades and blinks were excluded from the eye movement data and only the gaze points classified as fixation points were used. For each frame of the video sequence, the corresponding fixation points were processed as follows. First, the right-eye fixation points were shifted horizontally according to the right-to-left disparity map. Then, these points were combined with the left-eye fixation points and filtered with a Gaussian kernel to account for the eye tracking inaccuracies and the reduction of the visual sensitivity, which depends on the distance from the fovea. The standard deviation of the Gaussian filter used for computing the FDMs was set to 1 degree of visual angle, which corresponds to 60 pixels in our experiments. This is based on the assumption that the fovea of the human eye covers approximately 2 degrees of visual angle . Therefore, for each frame of the stereoscopic video sequence, only one FDM, corresponding to the left view, was produced from the left and right eye movements.
You can download all lists of fixation points and fixation density maps from the following FTP (please use dedicated FTP clients, such as FileZilla or FireFTP):
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.