Unsupervised Domain Adaptation Learning for Hierarchical Infant Pose Recognition with Synthetic Data

TL;DR

This paper introduces an unsupervised domain adaptation approach for hierarchical infant pose recognition that effectively utilizes synthetic data to improve accuracy on real infant images, aiding early developmental assessment.

Contribution

It proposes a CNN-based model with domain adaptation and hierarchical recognition, along with a new dataset, to enhance infant pose recognition accuracy across synthetic and real images.

Findings

Significant domain alignment between synthetic and real infant images.

Improved accuracy in fine-grained infant pose recognition.

Effective use of a new labeled dataset for training and evaluation.

Abstract

The Alberta Infant Motor Scale (AIMS) is a well-known assessment scheme that evaluates the gross motor development of infants by recording the number of specific poses achieved. With the aid of the image-based pose recognition model, the AIMS evaluation procedure can be shortened and automated, providing early diagnosis or indicator of potential developmental disorder. Due to limited public infant-related datasets, many works use the SMIL-based method to generate synthetic infant images for training. However, this domain mismatch between real and synthetic training samples often leads to performance degradation during inference. In this paper, we present a CNN-based model which takes any infant image as input and predicts the coarse and fine-level pose labels. The model consists of an image branch and a pose branch, which respectively generates the coarse-level logits facilitated by the unsupervised domain adaptation and the 3D keypoints using the HRNet with SMPLify optimization. Then the outputs of these branches will be sent into the hierarchical pose recognition module to estimate the fine-level pose labels. We also collect and label a new AIMS dataset, which contains 750 real and 4000 synthetic infants images with AIMS pose labels. Our experimental results show that the proposed method can significantly align the distribution of synthetic and real-world datasets, thus achieving accurate performance on fine-grained infant pose recognition.