Three-Dimensional Posture Estimation of Vehicle Occupants Using Depth and Infrared Images.

Pose estimation is crucial for ensuring passenger safety and better user experiences in semi- and fully autonomous vehicles. Traditional methods relying on pose estimation from regular color images face significant challenges due to a lack of three-dimensional (3D) information and the sensitivity to occlusion and lighting conditions. Depth images, which are invariant to lighting issues and provide 3D information about the scene, offer a promising alternative. However, there is a lack of strong work in 3D pose estimation from such images due to the time-consuming process of annotating depth images with 3D postures. In this paper, we present a novel approach to 3D human posture estimation using depth and infrared (IR) images. Our method leverages a three-stage fine-tuning process involving simulation data, approximated data, and a limited set of manually annotated samples. This approach allows us to effectively train a model capable of accurate 3D pose estimation with a median error of under 10 cm across all joints, using fewer than 100 manually annotated samples. To the best of our knowledge, this is the first work focusing on vehicle occupant posture detection utilizing only depth and IR data. Our results demonstrate the feasibility and efficacy of this approach, paving the way for enhanced passenger safety in autonomous vehicle systems.

Efficient Model-Based Anthropometry under Clothing Using Low-Cost Depth Sensors.

Measuring human body dimensions is critical for many engineering and product design domains. Nonetheless, acquiring body dimension data for populations using typical anthropometric methods poses challenges due to the time-consuming nature of manual methods. The measurement process for three-dimensional (3D) whole-body scanning can be much faster, but 3D scanning typically requires subjects to change into tight-fitting clothing, which increases time and cost and introduces privacy concerns. To address these and other issues in current anthropometry techniques, a measurement system was developed based on portable, low-cost depth cameras. Point-cloud data from the sensors are fit using a model-based method, Inscribed Fitting, which finds the most likely body shape in the statistical body shape space and providing accurate estimates of body characteristics. To evaluate the system, 144 young adults were measured manually and with two levels of military ensembles using the system. The results showed that the prediction accuracy for the clothed scans remained at a similar level to the accuracy for the minimally clad scans. This approach will enable rapid measurement of clothed populations with reduced time compared to manual and typical scan-based methods.

Longitudinal Degradation of Pavement Marking Detectability for Mobile LiDAR Sensing Technology in Real-World Use.

Recent advancements in vehicle automation and driver-assistance systems that detect pavement markings has increased the importance of the detectability of pavement markings through various sensor modalities across weather and road conditions. Among the sensing techniques, light detection and ranging (LiDAR) sensors have become popular for vehicle-automation applications. This study used low-cost mobile multi-beam LiDAR to assess the performance of several types of pavement marking materials installed on a limited-access highway in various conditions, and quantified the degradation in detection performance over three years. Four marking materials, HPS-8, polyurea, cold plastic, and sprayable thermoplastic, were analyzed in the current study. LiDAR reflectivity data extracted from a total of 210 passes through the test sections were analyzed. A new detectability score based on LiDAR intensity data was proposed to quantify the marking detectability. The results showed that the pavement marking detectability varied across the material types over the years. The results provide guidance for selecting materials and developing maintenance schedules when marking detectability by LiDAR is a concern.

A parametric modeling of adult body shape in a supported seated posture including effects of age.

Statistical body shape models (SBSM) provide compact, flexible representations of body shape that can be implemented in design software. However, few SBSMs have been created to represent adults in supported seated postures that are relevant for the design of seated environments, and none has incorporated the effects of age. This paper presents an SBSM based on surface laser-scan data from 155 U.S. adults. The data were processed to obtain homologous mesh structure and symmetric geometry, and the processed data were statistically analysed using principal component analysis to obtain a compact representation of the data variance. Regression analysis was conducted to predict body size and shape from stature, body mass index, ratio of sitting height to stature, sex, and age. The resulting model allows rapid generation of realistic body models for applications, including product design, accommodation assessment, and safety system optimisation. The model is publicly accessible at HumanShape.org. Practitioner summary: This paper presents a statistical model that represents adult body shapes in a supported seated posture based on 3 D anthropometric measurements. This model is the first whole-body parametric model known to incorporate age effects based on data extending beyond 65 years of age.

Static, Dynamic, and Cognitive Fit of Exosystems for the Human Operator.

OBJECTIVE: To define static, dynamic, and cognitive fit and their interactions as they pertain to exosystems and to document open research needs in using these fit characteristics to inform exosystem design. BACKGROUND: Initial exosystem sizing and fit evaluations are currently based on scalar anthropometric dimensions and subjective assessments. As fit depends on ongoing interactions related to task setting and user, attempts to tailor equipment have limitations when optimizing for this limited fit definition. METHOD: A targeted literature review was conducted to inform a conceptual framework defining three characteristics of exosystem fit: static, dynamic, and cognitive. Details are provided on the importance of differentiating fit characteristics for developing exosystems. RESULTS: Static fit considers alignment between human and equipment and requires understanding anthropometric characteristics of target users and geometric equipment features. Dynamic fit assesses how the human and equipment move and interact with each other, with a focus on the relative alignment between the two systems. Cognitive fit considers the stages of human-information processing, including somatosensation, executive function, and motor selection. Human cognitive capabilities should remain available to process task- and stimulus-related information in the presence of an exosystem. Dynamic and cognitive fit are operationalized in a task-specific manner, while static fit can be considered for predefined postures. CONCLUSION: A deeper understanding of how an exosystem fits an individual is needed to ensure good human-system performance. Development of methods for evaluating different fit characteristics is necessary. APPLICATION: Methods are presented to inform exosystem evaluation across physical and cognitive characteristics.

Parametric body shape model of standing children aged 3-11 years.

A statistical body shape model (SBSM) for children was developed for generating a child body shape with desired anthropometric parameters. A standardised template mesh was fit to whole-body laser scan data from 137 children aged 3-11 years. The mesh coordinates along with a set of surface landmarks and 27 manually measured anthropometric variables were analysed using principal component (PC) analysis. PC scores were associated with anthropometric predictors such as stature, body mass index (BMI) and ratio of erect sitting height to stature (SHS) using a regression model. When the original scan data were compared with the predictions of the SBSM using each subject's stature, BMI and SHS, the mean absolute error was 10.4 ± 5.8 mm, and 95th percentile error was 24.0 ± 18.5 mm. The model, publicly available online, will have utility for a wide range of applications. Practitioner Summary: A statistical body shape model for children helps to account for inter-individual variability in body shapes as well as anthropometric dimensions. This parametric modelling approach is useful for reliable prediction of the body shape of a specific child with a few given predictors such as stature, body mass index and age.

Child body shape measurement using depth cameras and a statistical body shape model.

We present a new method for rapidly measuring child body shapes from noisy, incomplete data captured from low-cost depth cameras. This method fits the data using a statistical body shape model (SBSM) to find a complete avatar in the realistic body shape space. The method also predicts a set of standard anthropometric data for a specific subject without measuring dimensions directly from the fitted model. Since the SBSM was developed using principal component (PC) analysis, we formulate an optimisation problem to fit the model in which the degrees of freedom are defined in PC-score space. The mean unsigned distance between the fitted-model based on depth-camera data and the high-resolution laser scan data was 9.4 mm with a standard deviation (SD) of 5.1 mm. For the torso, the mean distance was 2.9 mm (SD 1.4 mm). The correlations between standard anthropometric dimensions predicted by the SBSM and manually measured dimensions exceeded 0.9.

A three-dimensional parametric adult head model with representation of scalp shape variability under hair.

Modeling the shape of the scalp and face is essential for the design of protective helmets and other head-borne equipment. However, head anthropometry studies using optical scanning rarely capture scalp shape because of hair interference. Data on scalp shape is available from bald men, but female data are generally not available. To address this issue, scalp shape was digitized in an ethnically diverse sample of 100 adult women, age 18-59, under a protocol that included whole head surface scanning and scalp measurement using a three-dimensional (3D) coordinate digitizer. A combined male and female sample was created by adding 3D surface scans of a similarly diverse sample of 80 bald men. A statistical head shape model was created by standardizing the head scan data. A total of 58 anatomical head landmarks and 12 head dimensions were obtained from each scan and processed along with the scans. A parametric model accounting for the variability of the head shape under the hair as a function of selected head dimensions was developed. The full-variable model has a mean shape error of 3.8 mm; the 95th percentile error was 7.4 mm, which were measured at the vertices. The model will be particularly useful for generating a series of representing a target population as well as for generating subject-specific head shapes along with predicted landmarks and dimensions. The model is publicly available online at http://humanshape.org/head/.

A parametric model of child body shape in seated postures.

OBJECTIVE: The shape of the current physical and computational surrogates of children used for restraint system assessments is based largely on standard anthropometric dimensions. These scalar dimensions provide valuable information on the overall size of the individual but do not provide good guidance on shape or posture. This study introduced the development of a parametric model that statistically predicts individual child body shapes in seated postures with a few given parameters. METHODS: Surface geometry data from a laser scanner of children ages 3 to 11 (n = 135) were standardized by a 2-level fitting method using intermediate templates. The standardized data were analyzed by principal component analysis (PCA) to efficiently describe the body shape variance. Parameters such as stature, body mass index, erect sitting height, and 2 posture variables related to torso recline and lumbar spine flexion were associated with the PCA model using regression. RESULTS: When the original scan data were compared with the predictions of the model using the given subject dimensions, the average root mean square error for the torso was 9.5 mm, and the 95th percentile error was 17.35 mm. CONCLUSIONS: For the first time, a statistical model of child body shapes in seated postures is available. This parametric model allows the generation of an infinite number of virtual children spanning a wide range of body sizes and postures. The results have broad applicability in product design and safety analysis. Future work is needed to improve the representation of hands and feet and to extend the age range of the model. The model presented in this article is publicly available online through HumanShape.org.

A statistical skull geometry model for children 0-3 years old.

Head injury is the leading cause of fatality and long-term disability for children. Pediatric heads change rapidly in both size and shape during growth, especially for children under 3 years old (YO). To accurately assess the head injury risks for children, it is necessary to understand the geometry of the pediatric head and how morphologic features influence injury causation within the 0-3 YO population. In this study, head CT scans from fifty-six 0-3 YO children were used to develop a statistical model of pediatric skull geometry. Geometric features important for injury prediction, including skull size and shape, skull thickness and suture width, along with their variations among the sample population, were quantified through a series of image and statistical analyses. The size and shape of the pediatric skull change significantly with age and head circumference. The skull thickness and suture width vary with age, head circumference and location, which will have important effects on skull stiffness and injury prediction. The statistical geometry model developed in this study can provide a geometrical basis for future development of child anthropomorphic test devices and pediatric head finite element models.