TABLE I. Mean Values of Each Approach for Distance Estimation Error, CPU Usage, Battery Usage, and Field-of-view.
| Approach | Central Distance Error (cm) | L/R Peripheral Distance Error (cm) lm | CPU Usage | Battery Usage (%) | Field-of- | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| l m | 2 m | 3 m | 2 m | 3 m | (%) | lOmin | 30min | 60 min | View (°) | ||
| CoreML | 1.66 | 6.22 | 5.18 | (7.16.9.31) | (64.99,34.22) | (83.29,75.57) | 44 | 4 | 13 | 25 | 52 |
| IR_self | 1.30 | 1.48 | 1.96 | (15.88,25.35) | (17.76,39.36) | (65.94,35.36) | 50 | 3 | 10 | 21 | 40 |
| Li DAR_back | 1.36 | 0.87 | 0.81 | (9.66.6.68) | (18.77.8.65) | (27.87.25.47) | 48 | 7 | 20 | 40 | 40 |
| ARKit_self | 0.86 | 1.13 | 2.44 | (4.81.9.88) | (13.04.23.65) | (28.54.30.23) | 62 | 6 | 20 | 37 | 35 |
| ARKit_back | 1.37 | 0.48 | 1.40 | (7.08,4.75) | (1.88,23.73) | (7.21,20.78) | 74 | 5 | 14 | 27 | 30 |
Central and Peripheral Distance Errors Refer to Absolute Error Values in cm Relative to the Ground Truth Distance (1 m, 2 m, 3 m). Underlined Values Indicate That the Mean Values Were Underestimations While Regular Values Indicate overestimations. Bolded Values Indicate Best-in-Class Performance for That Specific Metric.