Table 2 Summary statistics for each data-set, comparing the size, resolution and demographics between the Oxfordshire and Sichuan study.
From: Reducing annotation burden in physical activity research using vision language models
Oxfordshire | Sichuan | |
|---|---|---|
Number of participants | 161 | 111 |
Number of labelled images (% all images) | 231,837 (74%) | 46,184 (34%) |
Median \(\delta t\) (1st, 3rd quartile) between images (s) | 24 (23, 32) | 84 (69, 88) |
No. unique labels | 220 | 110 |
Median instances per participant: Sedentary | 884 | 184 |
LIPA | 441.5 | 142 |
MVPA | 81 | 45 |
Number of participants (%) aged: 0–30 | 45 (28%) | 12 (11%) |
30–50 | 67 (42%) | 43 (41%) |
50–70 | 39 (24%) | 49 (47%) |
70–100 | 8 (6%) | 1 (1%) |
Sex: Female | 103 (64%) | 63 (58%) |
Male | 58 (36%) | 45 (42%) |
