Learning an Infant Body Model from RGB-D Data for Accurate Full Body Motion Analysis

Hesse, Nikolas; Pujades, Sergi; Romero, Javier; Black, Michael J.; Bodensteiner, Christoph; Arens, Michael; Hofmann, Ulrich G.; Tacke, Uta; Hadders‐Algra, Mijna; Weinberger, Raphael; Müller‐Felber, Wolfgang; Schröeder, Andreas

doi:10.1007/978-3-030-00928-1_89

Cited by 56 publications

(51 citation statements)

References 20 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In an effort to solve this [20] proposed a method for the generation of a synthetic dataset. They have produced the Moving INfants In RGB-D (MINI-RGBD) dataset by using their previously developed Skinned Multi-Infant Linear body model (SMIL) [21]. This dataset maps real infant movements to the SMIL model to generate anonymised, synthetic footage.…”

Section: A Synthetic Datasetmentioning

confidence: 99%

Establishing Pose Based Features Using Histograms for the Detection of Abnormal Infant Movements

McCay

Marcroft

et al. 2019

2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

View full text Add to dashboard Cite

The pursuit of early diagnosis of cerebral palsy has been an active research area with some very promising results using tools such as the General Movements Assessment (GMA). In this paper, we conducted a pilot study on extracting important information from video sequences to classify the body movement into two categories, normal and abnormal, and compared the results provided by an independent expert reviewer based on GMA. We present two new pose-based features, Histograms of Joint Orientation 2D (HOJO2D) and Histograms of Joint Displacement 2D (HOJD2D), for the pose-based analysis and classification of infant body movement from video footage. We extract the 2D skeletal joint locations from 2D RGB images using Cao et al.'s method [1]. Using the MINI-RGBD dataset [2], we further segment the body into local regions to extract part specific features. As a result, the pose and the degree of displacement are represented by histograms of normalised data. To demonstrate the effectiveness of the proposed features, we trained several classifiers using combinations of HOJO2D and HOJD2D features and conducted a series of experiments to classify the body movement into categories. The classification algorithms used included k-Nearest Neighbour (kNN, k=1 and k=3), Linear Discriminant Analysis (LDA) and the Ensemble classifier. Encouraging results were attained, with high accuracy (91.67%) obtained using the Ensemble classifier.

show abstract

Section: A Synthetic Datasetmentioning

confidence: 99%

Establishing Pose Based Features Using Histograms for the Detection of Abnormal Infant Movements

McCay

Marcroft

et al. 2019

2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

View full text Add to dashboard Cite

show abstract

“…The extraction of limb movement from optic flow also requires careful parameter selection and manual adjustment of the tracking algorithm. As an alternative, markerless methods have been developed for the tracking of infant movement [23], [24]. However, these methods require depth images, are computationally intensive and the models are likely to be overfit to the relatively small datasets on which they are trained.…”

Section: Introductionmentioning

confidence: 99%

Computer Vision to Automatically Assess Infant Neuromotor Risk

Chambers

Seethapathi

Saluja

et al. 2020

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

An infant's risk of developing neuromotor impairment is primarily assessed through visual examination by specialized clinicians. Therefore, many infants at risk for impairment go undetected, particularly in under-resourced environments. There is thus a need to develop automated, clinical assessments based on quantitative measures from widely-available sources, such as videos recorded on a mobile device. Here, we automatically extract body poses and movement kinematics from the videos of at-risk infants (N=19). For each infant, we calculate how much they deviate from a group of healthy infants (N=85 online videos) using a Naïve Gaussian Bayesian Surprise metric. After pre-registering our Bayesian Surprise calculations, we find that infants who are at high risk for impairments deviate considerably from the healthy group. Our simple method, provided as an open-source toolkit, thus shows promise as the basis for an automated and low-cost assessment of risk based on video recordings.

show abstract

“…The extraction of limb movement from optic flow also requires careful parameter selection and manual adjustment of the tracking algorithm. As an alternative, marker-less tracking methods have been developed for the tracking of infant movement (Hesse et al, 2018;Olsen, Herskind, Nielsen, & Paulsen, 2014). However, these methods require depth images, are computationally intensive, and the models are likely to be overfit to the relatively small datasets on which they are trained.…”

Section: Introductionmentioning

confidence: 99%

Computer vision to automatically assess infant neuromotor risk

Chambers

Seethapathi

Saluja

et al. 2019

Preprint

View full text Add to dashboard Cite

An infant's risk of developing neuromotor impairment is primarily assessed through visual examination by specialized clinicians. Therefore, many infants at risk for impairment go undetected, particularly in under-resourced environments. There is thus a need to develop automated, clinical assessments based on quantitative measures from widely-available sources, such as video cameras. Here, we automatically extract body poses and movement kinematics from the videos of at-risk infants (N=19). For each infant, we calculate how much they deviate from a group of healthy infants (N=85 online videos) using Naïve Gaussian Bayesian Surprise. After pre-registering our Bayesian Surprise calculations, we find that infants that are at higher risk for impairments deviate considerably from the healthy group. Our simple method, provided as an open source toolkit, thus shows promise as the basis for an automated and low-cost assessment of risk based on video recordings.

show abstract

Learning an Infant Body Model from RGB-D Data for Accurate Full Body Motion Analysis

Cited by 56 publications

References 20 publications

Establishing Pose Based Features Using Histograms for the Detection of Abnormal Infant Movements

Establishing Pose Based Features Using Histograms for the Detection of Abnormal Infant Movements

Computer Vision to Automatically Assess Infant Neuromotor Risk

Computer vision to automatically assess infant neuromotor risk

Contact Info

Product

Resources

About