UDC 519.68:681.513.7:591.169 Gaze Tracking Acceleration using CUDA Technology E. A. Sibirtseva, I. M. Gostev National Research University Higher School of Economic Department of Computer Science 20, Myasnitskaya, Moscow, Russia, 101000 Low-cost gaze tracking systems are in great demand due to their wide range of application. <...> Commonly, extra devices are needed (for instance, head mounted cameras); however, in this investigation gaze tracking is performed in real-time based on the video stream from an infrared video camera. <...> A methodology was developed which allows to calculate a gaze direction vector according to the relative position of eye center and corneal reflection from an infrared diode. <...> The centers of an eye and reflection are estimated using the vector field of image gradients and additional weighting. <...> The main advantage of the developed algorithm is the ability to detect and continuously track pupils’ centers, regardless of the head position, which significantly extends the scope of the gaze tracking system under consideration. <...> Key words and phrases: image processing, gaze-tracking, human-computer interaction, infrared illumination, CUDA, parallel computing; GPU; AHP. 1. <...> Introduction ment by using direct observation or special contact lens with a hole for pupil as in Edmund Huey research [1]. <...> Current studies into this field are mainly based on video recorded information and real-time analysis of gaze direction. <...> However, the use of natural eye movements in gaze-aware and attentive systems is beyond the scope of the current research. <...> However, tracking in real-time has a great limitation due to the nature of eye movements. <...> Sibirtseva E. A., Gostev I.M. Gaze Tracking Acceleration using CUDA. . . 69 are responsible for smoothly following a moving target. <...> Therefore, to track the eye movement it is needed to at least twice in 30 ms detects the gaze direction, while the saccade occurs. <...> Fortunately, this problem has a solution — processing not only on CPU (Central Processing Unit) but also on GPU (Graphics Processing Unit). <...> GPU consists of thousands of cores for efficient handling of concurrent tasks. <...> However there are several drawbacks in GPU computing: such as long time to data transfer from CPU to GPU <...>