BinParam: Binarized human parametric modeling via distribution alignment and orthogonal residuals

In this paper, a 3D human body meshed model with low resolution is obtained by parametric digital human modeling, but the meshed surface is not smooth enough to meet the requirements of practical application. To achieve the smoothing effect of the geometric model of digital human body, this paper proposes a mesh-smoothing method of 3D human model based on Sqrt3 subdivision algorithm. Experiment shows that this method applies to the multicurvature mesh body model based on parametric modeling method and achieves good results. The method is simple, efficient and of well-reserved feature.

Parametric 3D modeling of young women's lower bodies based on shape classification

The fusion of a human prior with neural rendering techniques has recently emerged as one of the most promising approaches to processing dynamic human‐body scenes with sparse inputs. However, learning geometric details and appearance in dynamic human‐body scenes based solely on a human prior model represents a severely under‐constrained problem. A new human‐body representation method to solve this problem: a neural radiance field with vertex‐texture latent codes (VT‐NeRF) is proposed. VT‐NeRF uses joint latent code to improve access to detailed information, combining vertex latent codes with 2D texture latent codes for the body surface. Referencing a 3D human skeleton for accurate guidance, the human model can quickly match poses and learn information about the body in different frames. VT‐NeRF can integrate body information from different frames and different poses quickly because it uses an information‐rich human prior: a 3D human skeleton and parametric models. A 3D human scene is then presented as an implied field of density and colour. Experiments with the ZJU‐MoCap dataset show that our method outperforms previous methods in terms of both novel‐view synthesis and 3D human reconstruction quality. It is twice as fast as Neural Body, and its average accuracy reaches 95.9%.

With the gradual increase in the demand for parametric human 3D models, the requirements for the accuracy of modeling and the quality of generation are getting higher and higher. In this paper, the human body structure model is constructed by 3D human posture estimation and spatio-temporal graph convolutional network, and after normalizing the data of the human body base, the 3D human body model based on the human body base is generated by combining the free deformation method, and the local light model is used to make the human body model more three-dimensional and real. In addition, the quality of 3D animation production is investigated by combining blind quality evaluation and validation analysis of the human body-based 3D model. The 3D human body modeling simulation error assessment has an error assessment root-means-square error that falls within 1.5%. In the validation analysis of model generation quality, the quality evaluation model proposed in this paper has a quality correlation coefficient and prediction accuracy of 0.9456 and 0.9454 on the dataset, respectively, and the performance of the ablation experiments and generalization experiments is stable. The mean value of the error between the animation production character generation effect and the actual testers is not more than 10mm.Combined with the research in this paper, it effectively improves the accuracy of human body modeling and provides an evaluation method for the quality of human 3D animation production and generation, which is of practical significance for its development.

This paper extends a prior human operator model to capture human steering performance in the teleoperation of unmanned ground vehicles (UGVs) in path-following scenarios with varying speed. A prior study presented a human operator model to predict human steering performance in the teleoperation of a passenger-sized UGV at constant speeds. To enable applications to varying speed scenarios, the model needs to be extended to incorporate speed control and be able to predict human performance under the effect of accelerations/decelerations and various time delays induced by the teleoperation setting. A strategy is also needed to parameterize the model without human subject data for a truly predictive capability. This paper adopts the ACT-R cognitive architecture and two-point steering model used in the previous work, and extends the model by incorporating a far-point speed control model to allow for varying speed. A parameterization strategy is proposed to find a robust set of parameters for each time delay to maximize steering performance. Human subject experiments are conducted to validate the model. Results show that the parameterized model can predict both the trend of average lane keeping error and its lowest value for human subjects under different time delays. The proposed model successfully extends the prior computational model to predict human steering behavior in a teleoperated UGV with varying speed. This computational model can be used to substitute for human operators in the process of development and testing of teleoperated UGV technologies and allows fully simulation-based development and studies.

Many current radar-based human detection systems employ some type of Doppler or Fourier-based processing, followed by spectrogram and gait analysis to classify detected targets. However, in Fourier-based techniques the maximum output signal-to-noise ratio (SNR) is given by targets whose target phase is linear. On the contrary, the phase variation of the human target response is nonlinear. This difference causes a significant loss in SNR, and therefore detection performance. In this paper, two novel, nonlinear phase detector designs based on human modeling are presented. In the first method, only the human torso reflections are modeled and unknown model parameters computed using Maximum Likelihood Estimation. In the second method, the entire human body is modeled as a different parametric model. The expected radar response for each combination of parameter values is stored in a database. An optimal sparse approximation to the data is found using Orthogonal Matching Pursuit. The performance of the proposed techniques and optimal space-time adaptive processing algorithm is compared and target characterization applications examined.

Multi-view human pose and shape estimation via mesh-aligned voxel interpolation

This paper aims to build the human body models based on the belly shape features. We use NURBS surface to build the human body models, and control vertex in the surface to deform. Based on the OpenGL, the integral deformation of the height and width of the whole human body model was carried out. Specifically, girth, radial and bulging deformation was accomplished on the area of belly. Finally, we get the human body models with different belly shape, which is of significance.

<div class="section abstract"><div class="htmlview paragraph">This study concerns the generation of response surfaces for kinematics and injury prediction in pedestrian impact simulations using human body model. A 1000-case DOE (Design of Experiments) study with a Latin Hypercube sampling scheme is conducted using a finite element pedestrian human body model and a simplified parametric vehicle front-end model. The Kriging method is taken as the approach to construct global approximations to system behavior based on results calculated at various points in the design space. Using the response surface models, human lower limb kinematics and injuries, including impact posture, lateral bending angle, ligament elongation and bone fractures, can be quickly assessed when either the structural dimensions or the structural behavior of the vehicle front-end design change. This will aid in vehicle front-end design to enhance protection of pedestrian lower limbs.</div></div>

We propose a markerless end-to-end training framework for parametric 3D human shape models. The training of statistical 3D human shape models with minimal supervision is an important problem in computer vision. Contrary to prior work, the whole training process (i) uses a differentiable shape model surface and (ii) is trained end-to-end by jointly optimizing all parameters of a single, self-contained objective that can be solved with slightly modified off-the-shelf non-linear least squares solvers. The training process only requires a compact model definition and an off-the-shelf 2D RGB pose estimator. No pre-trained shape models are required. For training (iii) a medium-sized dataset of approximately 1000 low-resolution human body scans is sufficient to achieve competitive performance on the challenging FAUST surface correspondence benchmark. The training and evaluation code will be made available for research purposes to facilitate end-to-end shape model training on novel datasets with minimal setup cost.

Bottom-up in vitro ADME parameterized PBK models play a vital role in Next Generation Risk Assessment (NGRA), which evaluates the toxicological hazards of compounds through New Approach Methods (NAMs). The ZeroPM project, funded by the EU under Horizon 2020, develops strategies to prioritize persistent (P) and mobile (M) compounds for regulatory measures based on exposure levels and toxicological properties. This study developed a quantitative in vitro to in vivo extrapolation (QIVIVE) approach for two classes of PM compounds - 9 triazines and 16 triazoles. The virtual in vitro distribution model (VIVD) derived free medium benchmark concentration values (EC20 and IR1.5 (mol/L)) from a previously published NAMs testing battery. The VIVD model accounted for partitioning and ionization of PM compounds, while a bottom-up PBK model using in vitro ADME data accurately predicted rat plasma concentrations for tebuconazole and cyromazine. The same assumptions were translated to a human oral-route PBK model. Most triazines/triazoles had in vitro ADME data, where the data gaps were filled using read-across approach. The human PBK model employed forward dosimetry to estimate plasma concentrations from preclinical lowest observed adverse effect levels (LOAELs) or hypothetical exposure data, using surface- and groundwater concentrations as worst-case scenarios. Comparison to QIVIVE data indicates that NAMs-derived values are protective and can serve as basis for human risk assessment or risk-based prioritization.

Photo-realistic modeling and rendering of humans is extremely important for virtual reality (VR) environments, as the human body and face are highly complex and exhibit large shape variability but also, especially, as humans are extremely sensitive to looking at humans. Further, in VR environments, interactivity plays an important role. While purely computer graphics modeling can achieve highly realistic human models, achieving real photo-realism with these models is computationally extremely expensive. In this chapter, a full end-to-end pipeline for the creation of hybrid representations for human bodies and faces (animatable volumetric video) is investigated, combining classical computer graphics models with image- and video-based as well as example-based approaches: by enriching volumetric video with semantics and animation properties and applying new hybrid geometry- and video-based animation methods, we bring volumetric video to life and combine interactivity with photo-realism. Semantic enrichment and geometric animation ability is achieved by establishing temporal consistency in the 3D data, followed by an automatic rigging of each frame using a parametric shape-adaptive full human body model. For pose editing, we exploit the captured data as much as possible and kinematically deform selected captured frames to fit a desired pose. Further, we treat the face differently from the body in a hybrid geometry- and video-based animation approach, where coarse movements and poses are modeled in the geometry, while very fine and subtle details in the face, often lacking in purely geometric methods, are captured in video-based textures. These are processed to be interactively combined to form new facial expressions. On top of that, we learn the appearance of regions that are challenging to synthesize, such as the teeth or the eyes, and fill in missing regions realistically with an autoencoder-based approach.KeywordsVolumetric videoAnimationHuman modelingHybrid modeling

Recently, implicit neural representation has been widely used to learn the appearance of human bodies in the canonical space, which can be further animated using a parametric human model. However, how to decompose the material properties from the implicit representation for relighting has not yet been investigated thoroughly. We propose to address this problem with a novel framework, ReN Human, that takes sparse or even monocular input videos collected in unconstrained lighting to produce a 3D human representation that can be rendered with novel views, poses, and lighting. Our method represents humans as deformable implicit neural representation and decomposes the geometry, material of humans as well as environment illumination for capturing a relightable and animatable human model. Moreover, we introduce a volumetric lighting grid consisting of spherical Gaussian mixtures to learn the spatially varying illumination and animatable visibility probes to model the dynamic self-occlusion caused by human motion. Specifically, we learn the material property fields and illumination using a physically-based rendering layer that uses Monte Carlo importance sampling to facilitate differentiation of the complex rendering integral. We demonstrate that our approach outperforms recent novel views and poses synthesis methods in a challenging benchmark with sparse videos, enabling high-fidelity human relighting.

ABSTRACTHuman mesh reconstruction from point clouds has gained increasing attention in recent years. However, existing reconstruction pipelines predominantly focus on regressing human model parameters, which results in an indirect mapping between input observations and output meshes. In this letter, we introduce a point‐to‐vertex reconstruction method, which bypasses parameter spaces to reconstruct complete SMPL topology from partial point observations directly. Specifically, our framework employs a hierarchical approach to extract geometric information from the partial point cloud. The feature mapping process is realized via a transformer‐based architecture, which integrates parametric human model priors and employs hybrid position encoding. To further improve accuracy, the dual refinement module progressively refines the reconstruction process through sequential optimization in the coordinate and vertex‐level feature spaces. Experiments on SURREAL datasets demonstrate that our framework achieves state‐of‐the‐art performance, surpassing previous methods by 18.4% in MPVE (from 19.6 to 16.0 mm) and 10.2% in MPJPE (from 17.7 to 15.9 mm).

Design of robotic mannequin formed by flexible belt net