| 11 | | After doing a performance evaluation it was shown that Pupil can provide an average gaze estimation accuracy of 0.6 degree of visual angle (0.08 degree precision) with a processing pipeline latency of only 0.045 seconds. |
| 12 | | |
| 13 | | |
| 14 | | In order to know what someone is looking at, we must to establish a mapping between pupil and gaze positions. This is what we call calibration. The following pictures will show us the process in using Pupil hardware and software for gaze-tracking. |
| 15 | | |
| 16 | | [[Image(p1.png,30%)]] |
| 17 | | |
| 18 | | **Pic.1 Positioning of the headset and adjusting the direction of the gaze** |
| 19 | | |
| 20 | | [[Image(p2.png)]] |
| 21 | | |
| 22 | | **Pic.2 A correct way of looking at the camera (1) and three following incorrect was of doing that (2,3,4)** |
| 23 | | |
| 24 | | [[Image(p3.png)]] |
| 25 | | |
| 26 | | **Pic.3 Points for calibrating our gaze** |
| 27 | | |
| 28 | | [[Image(p4.png,30%)]] |
| 29 | | |
| 30 | | **Pic.4 The process of recording the gaze** |
| 31 | | |
| 32 | | [[Image(p5.png)]] |
| 33 | | |
| 34 | | **Pic.5 Recording live gaze example** |
| 35 | | |
| 36 | | [[Image(p6.jpg)]] |
| 37 | | |
| 38 | | **Pic.6 Information gained about the subjects gaze after recording** |
