Real-time human pose estimation on a smart walker using convolutional neural networks. (1st December 2021)

• Record Type: Journal Article
• Authors: Palermo, Manuel; Moccia, Sara; Migliorelli, Lucia; Frontoni, Emanuele; Santos, Cristina P.
• Is Part Of: Expert systems with applications. Volume 184(2021)
• Publication Date: 2021-12-01
• Subjects: rehabilitation -- smart walker -- computer vision -- deep learning -- human pose estimation; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2021.115498
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: Real-time full-body human pose estimation solution for the ASBGo smart walker. Convolutional neural network for 2D keypoint detection with regression to 3D space. Information fusion from two rgb+d cameras with non-overlapping, complementary views. Exploration of data acquisition, model training, benchmarking and deployment on CPU. Abstract: Rehabilitation is important to improve quality of life for mobility-impaired patients. Smart walkers are a commonly used solution that should embed automatic and objective tools for data-driven human-in-the-loop control and monitoring. However, present solutions focus on extracting few specific metrics from dedicated sensors with no unified full-body approach. We investigate a general, real-time, full-body pose estimation framework based on two RGB+D camera streams with non-overlapping views mounted on a smart walker equipment used in rehabilitation. Human keypoint estimation is performed using a two-stage neural network framework. The 2D-Stage implements a detection module that locates body keypoints in the 2D image frames. The 3D-Stage implements a regression module that lifts and relates the detected keypoints in both cameras to the 3D space relative to the walker. Model predictions are low-pass filtered to improve temporal consistency. A custom acquisition method was used to obtain a dataset, with 14 healthy subjects, used for training and evaluating the proposed framework offline, which was then deployed on the real walker equipment. An overall keypoint detection error of 3.73 pixels for the 2D-Stage and 44.05 mm for the 3D-Stage were reported, with an inference time of 26.6 ms when deployed on the constrained hardware of the walker. We present a novel approach to patient monitoring and data-driven human-in-the-loop control in the context of smart walkers. It is able to extract a complete and compact body representation in real-time and from inexpensive sensors, serving as a common base for downstream metrics extraction solutions, and Human-Robot interaction applications. Despite promising results, more data should be collected on users with impairments, to assess its performance as a rehabilitation tool in real-world scenarios.