The early diagnosis of cerebral palsy is an area which has recently seen significant multi-disciplinary research. Diagnostic tools such as the General Movements Assessment (GMA), have produced some very promising results. However, the prospect of automating these processes may improve accessibility of the assessment and also enhance the understanding of movement development of infants. Previous works have established the viability of using posebased features extracted from RGB video sequences to undertake classification of infant body movements based upon the GMA. In this paper, we propose a series of new and improved features, and a feature fusion pipeline for this classification task. We also introduce the RVI-38 dataset, a series of videos captured as part of routine clinical care. By utilising this challenging dataset we establish the robustness of several motion features for classification, subsequently informing the design of our proposed feature fusion framework based upon the GMA. We evaluate our proposed framework's classification performance using both the RVI-38 dataset and the publicly available MINI-RGBD dataset. We also implement several other methods from the literature for direct comparison using these two independent datasets. Our experimental results and feature analysis show that our proposed pose-based method performs well across both datasets. The proposed features afford us the opportunity to include finer detail than previous methods, and further model GMA specific body movements. These new features also allow us to take advantage of additional body-part specific information as a means of improving the overall classification performance, whilst retaining GMA relevant, interpretable, and shareable features.
In this paper, a system with six depth cameras was built to scan both feet simultaneously. An improved calibration method based on a T-shaped checkerboard was used to calculate the extrinsic parameters of the cameras. T-shaped virtual checkerboards were introduced to further fine-tune the accuracy of calibration based on the iterative closest point algorithm. Based on the proposed foot scanner, a complete procedure was introduced to measure the foot automatically by locating the anatomical landmarks without manual intervention. Various experiments were presented to validate the performance of the scanner and the measurements. The results verified that the proposed methods were efficient and versatile for three-dimensional foot scanning and measurement.
In this paper, a new method was proposed to establish the relationship between three-dimensional (3D) foot shapes and their two-dimensional (2D) foot silhouettes, through which a complete 3D foot shape can be predicted by simply inputting its two 2D silhouettes. 3D foot scans of 80 participants were randomly selected as the training set, and those of another 20 participants were used as the testing set. Elliptical Fourier analysis (EFA) and principle component analysis (PCA) were adopted to parameterize the 3D foot shapes. A linear regressive model was then developed to predict the 3D foot shape with the foot silhouettes. Experiment results indicated individual 3D foot shape can be predicted with a mean error between 1.21 and 1.27 mm, which can provide enough accuracy for the fit evaluation of footwear.
Purpose
The automatic body measurement is the key of tailoring, mass customization and fit/ease evaluation. The major challenges include finding the landmarks and extracting the sizes accurately. The purpose of this paper is to propose a new method of body measurement based on the loop structure.
Design/methodology/approach
The scanned human model is sliced equally to layers consist of various shapes of loops. The semantic feature analysis has been regarded as a problem of finding the points of interest (POI) and the loop of interest (LOI) according to the types of loop connections. Methods for determining the basic landmarks have been detailed.
Findings
The experimental results validate that the proposed methods can be used to locate the landmarks and to extract sizes on markless human scans robustly and efficiently.
Originality/value
With the method, the body measurement can be quickly performed with average errors around 0.5 cm. The results of segmentation, landmarking and body measurements also validate the robustness and efficiency of the proposed methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.