Webpreterm infants (the babyPose dataset). Results: When ap-plied to pose estimation, the median root mean square dis-tance, computed among all limbs, between the estimated and the ground-truth pose was 9.06 pixels, overcoming approaches based on spatial features only (11.27 pixels). Conclusion: Results showed that the spatio-temporal fea- WebOct 15, 2024 · Dataset. The dataset used in this work expanded the babyPose dataset presented in Migliorelli et al. (2024) by adding 11 depth videos, for a total of 27 depth videos. All videos were acquired in the NICU of the G. Salesi Hospital in Ancona, Italy. The infants were identified by clinicians in the NICU among those who were spontaneously breathing.
(PDF) Supine Infant Pose Estimation via Single Depth Image
WebJun 6, 2024 · Results: When applied to pose estimation, the median root mean square distance (RMSD) calculated among all limbs is 4.12 pixels, which is much lower than the … Web10 rows · Dec 1, 2024 · The babyPose dataset was a collection of 16 depth videos of 16 preterm infants, acquired in the ... green light means in traffic
Preterm Infants
WebOct 15, 2024 · The babyPose dataset, which originally consisted of depth videos from 16 preterm infants, here counts 27 videos from 27 preterm infants. All the videos were acquired in the NICU of the “G. Salesi” Hospital in Ancona. Compatibly with the pandemic situation, we worked to increase the babyPose dataset size. WebThis dataset represents a unique resource for artificial intelligence researchers that want to develop algorithms to provide healthcare professionals working in NICUs with decision support. Hence, the babyPose dataset is the first annotated dataset of depth images relevant to preterm infants' movement analysis. Agid: 7252972 WebFig. 1. Depth-image acquisition setup. The depth camera (yellow boxes) is positioned at 40cm over the infant’s crib and does not hinder health-operator movements. An example of the acquired depth frame is showed on the top left corner (orange box). - … flying craftsman