Detection, tracking, and re-identification (ReID) of children wearing similar uniforms in a kindergarten environment is a very complex challenge for computer vision. Traditional surveillance systems or simple convolutional neural network (CNN) models often fail to distinguish children in crowds and occlusions. To address this challenge, this study proposes a novel hybrid framework combining YOLOv8 and Vision Transformer (ViT). Using YOLOv8 for detection and ViT for global feature extraction, we trained the model on a custom dataset of 31,521 images, achieving an overall accuracy of 93.75%, and the public benchmark MOT20 dataset of 28,630 images, achieving an overall accuracy of 96.02%. Our system showed remarkable success in tracking performance, where it achieved 86.7% MOTA and 99.7% IDF1 scores. This high IDF1 score proves that the model is highly effective in preventing identity switch. The main novelty of this study is the behavioral analysis of children beyond the boundaries of surveillance, where we measure walking distance and trajectory, and screen time. Finally, through cross-dataset comparison with the MOT20 public benchmark, we demonstrated that our proposed customized model is much more effective than current state-of-the-art methods in overcoming the domain gap in specific environments such as kindergarten.

The doubly labeled water (DLW, ²H₂¹⁸O) method for calculating the total production of CO₂ over several days is currently considered to be the most accurate technique for the measurement of total energy expenditure (TEE), and the results obtained using this method have been used to review energy requirements. Presently, there is limited data available on TEE in Japanese children. The objective of this study was to assess the TEE in pre-school Japanese children using the DLW method. We used a cross-sectional population of 140 children (69 boys and 71 girls) aged 3–6 years. TEE was measured using the DLW method over 8 days under free-living conditions. The average weights (kg) of the boys and girls were 15.6 ± 2.5 and 15.0 ± 2.1 for the 3–4 years old and 19.8 ± 3.8 and 19.6 ± 2.7 for the 5–6 years old, respectively. The corresponding TEE (kcal/day) was 1260.9 ± 357.8 and 1265.2 ± 408.0, and 1682.3 ± 489.0 and 1693.1 ± 473.3, respectively, showing a significant difference with respect to age. Furthermore, TEE per body weight (kcal/kg/day) was 83.2 ± 29.2 and 84.9 ± 26.6, and 85.4 ± 23.2 and 86.7 ± 22.6, respectively. However, when TEE was adjusted for body weight or fat-free mass, there were no age or sex differences. We conclude that in Japanese children, TEE in those aged 3–4 years was similar to the current Ministry of Health recommendations. However, TEE in children aged 5–6 years was slightly higher than the recommendations. Based on these findings, the present results obtained from a large number of participants will provide valuable reference data for Japanese children.