Metadata for publication "Comparison of marker-less 2D image-based methods for infant pose estimation"
Abstract
This study compares the performance of available generic- and specialized infant-pose estimators for a video-based automated general movement assessment (GMA), and the choice of viewing angle for optimal recordings, i.e., conventional diagonal view used in GMA vs. top-down view. 4500 annotated video-frames from 75 recordings of infant spontaneous motor functions from 4 to 16 weeks were used. The models retrained on infant data are published on Zenodo: https://doi.org/10.5281/zenodo.14833182
Fields of Study & Subjects
Human Pose and Action Recognition
Face recognition and analysis
Hand Gesture Recognition Systems
Publication details
- Journal / source
- Göttingen Research Online
- Publication date
- 20 January 2026
- Document type
- Dataset