An image of a user's eyes and face may be analyzed using computer-vision algorithms. A computing device may use the image to determine the location of the user's eyes and estimate the direction in which the user is looking. The eye tracking technology may be used in a wide range of lighting conditions and with many different and varying light levels. When a user is near a light source, an automatic exposure feature in the camera may result in the user's face and eyes appearing too dark in the image, possibly reducing the likelihood of face and eye detection. Adjusting attributes such as the camera exposure time and the intensity and illumination interval of the light sources based on motion and light levels may improve detection of a user's features.