Infant 2D Pose Estimation and Posture Detection System
AssigneeNortheastern University
InventorsSarah OSTADABBAS, Xiaofei HUANG, Nihang FU, Shuangjun LIU
Life of this patent
Estimating the term…
Abstract
Methods are provided for estimating a pose of an infant using image analysis and artificial intelligence. A classifier is trained using a dataset containing hybrid synthetic and real infant pose data. Multi-stage invariant representation machine learning strategies are employed that transfer knowledge from adjacent domains of adult poses and synthetic infant images into a fine-tuned domain-adapted infant pose estimation model.
Claims21
A system for estimating a pose of an infant, the system comprising a computer comprising one or more processors and memory and a pose estimator, the pose estimator comprising a trained model for estimating infant poses trained on an adult pose dataset and an augmented dataset containing real infant pose data and synthetic infant pose data using domain adaptation to align features of the synthetic infant pose data with the real infant pose data;
- and
- wherein the computer further comprises computer-readable instructions stored in the memory that when executed cause the pose estimator to:
- (a) receive one or more images of an infant using the one or more imaging devices;
- and
- (b) determine a pose of the infant using the trained model.
Dependencies are inferred from the claim text.