Name
Abstract | ||
|---|---|---|
Quality, diversity, and size of training data are critical factors for learning-based gaze estimators. We create two datasets satisfying these criteria for near-eye gaze estimation under infrared illumination: a synthetic dataset using anatomically-informed eye and face models with variations in face shape, gaze direction, pupil and iris, skin tone, and external conditions (2M images at 1280x960), and a real-world dataset collected with 35 subjects (2.5M images at 640x480). Using these datasets we train neural networks performing with sub-millisecond latency. Our gaze estimation network achieves 2.06(±0.44)° of accuracy across a wide 30°×40° field of view on real subjects excluded from training and 0.5° best-case accuracy (across the same FOV) when explicitly trained for one real subject. We also train a pupil localization network which achieves higher robustness than previous methods.
|
Year | DOI | Venue |
|---|---|---|
2019 | 10.1145/3290605.3300780 | CHI |
Keywords | Field | DocType |
dataset, eye tracking, machine learning, virtual reality | Computer vision,Gaze,Latency (engineering),Computer science,Pupil,Robustness (computer science),Eye tracking,Human–computer interaction,Artificial intelligence,Latency (engineering),Artificial neural network,Estimator | Conference |
ISBN | Citations | PageRank |
978-1-4503-5970-2 | 9 | 0.48 |
References | Authors | |
0 | 8 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Joohwan Kim | 1 | 9 | 2.17 |
| Michael Stengel | 2 | 120 | 9.80 |
| Alexander Majercik | 3 | 9 | 0.48 |
| Shalini Gupta | 4 | 299 | 20.42 |
| David Dunn | 5 | 30 | 2.78 |
| Samuli Laine | 6 | 1250 | 50.48 |
| Morgan Mcguire | 7 | 752 | 54.30 |
| David Luebke | 8 | 2196 | 140.84 |