Elsevier

Computer-Aided Design

Volume 39, Issue 7, July 2007, Pages 583-597
Computer-Aided Design

The effect of tendons on foot skin deformation

https://doi.org/10.1016/j.cad.2007.01.013Get rights and content

Abstract

Anatomical human models are usually divided into layers including skin, muscle and skeleton. In spite of the realistic animation of the models that can be achieved, and the realistic appearance of the model determined by the underlying muscles and skeleton, the role of tendons in determining the deformation of skin surface has not been well addressed.

This paper presents an approach for modeling human foot tendons and determines their influence on the skin layer deformation. Our goal is to model deformation of the tendons such that a realistic foot simulation can be obtained. An anatomical foot model including skin, muscle, tendon and skeleton layers is adopted. The appearance of the skin layer is determined based on the underlying layers. To allow interactive deformation of the tendon models, the axial deformation technique is adopted. Given the position of the foot and the basis function, the position of the data points that control the axial curve is updated. To allow more accurate computation of the data point positions, a method that estimates the basis function based on real data obtained from foot images is also presented. Experimental results showed that the axial deformation technique can model deformation of the foot tendons with satisfactory visual realism. With the tendon deformation, the visual realism on the skin deformation is also enhanced.

Introduction

Since the early 1970s, simulation of humans, and other animals has been one of the most important and interesting areas for researchers in CAD and computer graphics. However, due to the complexity of simulating the human body and its behavior, generating a realistic human model remains a difficult task.

To generate realistic human models or create animations, a direct approach is applying deformation on the skin surface of the characters. An accurate deformation of the skin layer can be modeled using physically based approaches including the finite element method (FEM). Recently, the development of more accurate modeling of the skin layer of human characters in computer graphics is based on their anatomical structure. The appearance of the skin is determined by the underlying muscles and skeleton layers. Although a substantial amount of work has been devoted to anatomically based modeling, deformation of the skin surface due to the underlying tendons has not been well addressed. Actually, the tendons play an important role in determining the skin surface deformation, especially on the foot and hand. For instances, the big toe pushing against the floor brings the tendon of the big toe into view (Fig. 1(a)) and the tendons appear on the surface of the hand when the fingers are extended (Fig. 1(b)).

Modeling tendon deformation is difficult because tendons may lie across several bone segments or have a large change in the direction of the line of action. One approach is to update the position of the tendon models by determining the movement of the corresponding action lines during motion. Modeling tendon motion according to muscle action lines is simple and easy to implement. However, the complicated motion of the tendon is largely simplified to move along a line segment. Besides, it is difficult to model the deformation of tendons lying across more than one skeleton joint. In this paper, a different approach for modeling tendon deformation is presented. The approach makes use of an axial curve which describes the deformation of tendon models. The complicated shapes of tendons are defined by the axial curve and used to control the deformation of the tendons during motion. A set of data points is used to define the shape of the axial curve. We present the tendon deformation on a human foot model. Our objective is to model deformation of the tendons such that a realistic human foot model can be simulated. The method is capable of achieving interactive deformation.

The tendon deformation technique is presented on a human foot model because:

  • 1.

    Foot tendons are one of the major underlying anatomical structures of the foot. They influence the appearance of the skin surface of the foot prominently;

  • 2.

    The foot contains many complex tendons to be modeled. The tendons may lie across several bone segments or have large changes in the direction of the line of action;

  • 3.

    Simulating foot tendon deformation has potential applications in shoe design. Fig. 2 shows a comparison of the foot model with and without tendons on wearing a slipper. The foot model with tendons has a larger contact region with the slipper. The simulation is useful for comfort analysis of shoes; and

  • 4.

    The technique for foot tendon deformation can be applied to other tendon deformation problems.

Over the past two decades, a great deals of effort has been focused on simulating human models [1]. However, it remains a great challenge in computer graphics because of the complexity of the human body and the animation of deformable skin layer of human characters requires geometric models changing over time. Free Form Deformation (FFD) introduced by Sederberg and Parry [2] provides flexibility for the users to control deformations of models. Barr [3] introduced the geometric method for modeling deformation of complex models. Although these methods provide flexibility for the users to control deformation, the realism of model deformation depends on the skill of the users rather than the intrinsic or physical properties of the models. In addition, the method is highly labour intensive and requires considerable skill in creating a realistic model. An accurate model of the skin layer of a character can also be obtained using physically based approaches such as FEM. Gourret et al. [4] simulate human skin deformation in a grasping task using FEM. The FEM is also applied to modeling the deformation of the underlying soft tissues or human organs [5], [6] and muscles [7], [8].

Recently, more accurate simulation of human characters has been developed based on their anatomical structure [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19]. The technique divides the character models into layers including skin, fatty tissue, muscle and skeleton. According to the motions of the characters, the skeleton is transformed and the muscles are deformed accordingly. The appearance of the skin layer is determined by applying deformation on the multi-layered models [9]. Chadwick [10] proposed a methodology for creating and animating layered characters. Muscles are deformed using a simple geometric modeling technique and are used to determine the resulting geometric surface of the character. Scheepers [11] introduced an anatomy-based approach for modeling human musculation. Wilhelms [12], [13] developed a muscle model based on a simple deformable cylinder. The skin is generated by voxelizing the underlying muscles, bones and tissues, which is deformed during joint motions. Nedel and Thalmann [14] proposed to create a multi-layered human model. The muscle layer is deformed based on physical concepts. Kähler et al. [15] presented and constructed a muscle model to animate facial expressions. The muscle model is simplified as a quadratic shape. The mass–spring system is applied to simulate the fatty subcutaneous layer and control deformation of the skin layer. Albrecht et al. [16] constructed and animated an anatomically based human hand model by employing pseudo-muscles that compute the finger movements based on anatomical data and mechanical laws. The geometric muscles deform the skin tissue using a mass–spring system. Physics-based facial animation techniques are proposed in [17], [18]. An anatomy-based approach for face modeling is presented in [19]. Skeletal subspace deformation (SSD) [20] is a common approach used in surface deformation for character animation. Nevertheless, the method may not produce the desired results in regions around complicated joints and the range of motion is restricted. Kry [21] addressed these problems by using eigendisplacements and improved the results. Lewis et al. [22] proposed the pose space deformation (PSD) technique for skeleton driven deformation. Despite the large amount of research on simulating character models, the role of tendons in determining deformation of the skin surface has not been well addressed.

Dong and Clapworthy [23] presented an anatomy-based approach for human modeling based on the simulation of muscles. They attempt to simulate realistic muscle deformations by transforming the muscles according to the action lines. A cubic curve is used to fit the action line that consists of several line segments. The positions of the tendon models are updated by maintaining their angles with the action lines. Although the role of tendon models has been considered and updated according to the action lines, a detailed discussion on tendon deformations (e.g. bending of the tendons) and the corresponding deformation of the skin layer has not been presented. The method of tendon deformation presented in this paper attempts to tackle this problem.

The remainder of this paper is organized as follows. In Section 2 we identify the anatomical structures of the foot and discuss the tendon anatomy. In Section 3 we describe our method for modeling foot tendons. A technique that computes the basis function based on foot images is described in Section 4. We present the experimental results for modeling foot tendon and skin deformation in Section 5. Section 6 concludes the paper.

Section snippets

Foot anatomy

The foot is a vital structure that connects between human body and the ground. It plays a highly important role in human locomotion because it supports and distributes the body’s weight. Zatsiorsky [24] described the kinematics of human motion. A detailed description of human foot anatomy can be found in [25], [26], [27].

The method

The foot model used in this research consists of four major layers including the muscles, tendons, skeleton and skin layers. Components in each layer are triangle meshes. Given the motion of the foot, the skeleton is transformed accordingly. Based on the position of the skeleton, the tendons and muscles are deformed. By establishing vertex correspondence between the skin and the underlying layers of muscles, tendons and skeleton, movements of muscles, tendons and skeleton can be used to

Determining the normalized basis function from images

To obtain an accurate foot tendon deformation, users are required to identify a suitable basis function for the deformation. The accuracy of deformation depends on the selection of the basis function. An alternative approach is to compute the basis function at the MTP joint based on the feature points identified from images. The technique is simple, flexible and requires only a low-cost camera for the image acquisition.

In our approach, the feature points are identified manually. We have

Implementation and results

An experimental system is implemented to visualize the tendon deformation when the toes are extended. These are the extensor digitorum longus and the extensor hallucis longus (Fig. 5, Fig. 6). We present our method by modeling the extensor digitorum longus and the extensor hallucis longus when the toes are extended. The influence of the tendon model on the deformation of the skin surface is also demonstrated. Besides, a comparison is made between the foot model with and without applying tendon

Conclusions

In this paper, a technique for foot tendon deformation is presented. Anatomically based modeling divides human characters into layers including skin, muscles and skeleton. Skin deformation due to the underlying tendons has not been well addressed. The focus of this paper is to model deformation of the tendons such that a realistic deformation of the skin layer and simulation of the foot model can be achieved. In the presented foot model, the skin layer is deformed according to the underlying

Acknowledgements

This work is partially supported by a grant from the Research Grants Council of the Hong Kong Special Administration Region (Project No. CUHK4197/04E) and an Innovation and Technology Fund from the Innovation and Technology Commission of the Hong Kong Special Administration Region (Project No. ITS/091/02).

References (39)

  • F. Lazarus et al.

    Axial deformations: An intuitive deformation technique

    Computer-Aided Design

    (1994)
  • K.C. Hui

    Free-form design using axial curve-pairs

    Journal of Computer-Aided Design

    (2002)
  • F. Klok

    Two moving coordinate frames for sweeping along a 3D trajectory

    Computer-Aided Geometric Design

    (1986)
  • A. Aubel et al.

    Realistic deformation of human body shapes

  • Sederberg TW, Parry SR. Free-form deformation of solid geometric models. In: SIGGRAPH, vol. 20(4). 1986. p....
  • Barr AH. Global and local deformations of solid primitives. In: Computer graphics (SIGGRAPH ’84 proceedings). p....
  • J.P. Gourret et al.

    Simulation of object and human skin deformations in a grasping task

    Computer Graphics

    (1989)
  • M. Bro-Nielsen et al.

    Real-time volumetric deformable models for surgery simulation using finite elements and condensation

    Computer Graphics Forum

    (1996)
  • S. Cotin et al.

    Real-time elastic deformations of soft tissues for surgery simulation

    IEEE Transactions on Visualization and Computer Graphics

    (1999)
  • Chen DT, Zelter D. Pump it up: Computer animation of a biomechanically based model of muscle using the finite element...
  • Q.H. Zhu et al.

    Real-time biomechanically-based muscle volume deformation using FEM

    Computer Graphics Forum

    (1998)
  • R. Turner et al.

    Interactive construction and animation of layered elastically deformable characters

    Computer Graphics Forum

    (1998)
  • J. Chadwick et al.

    Layered construction for deformable animated characters

    Computer Graphics

    (1989)
  • F. Scheepers et al.

    Anatomy-based modeling of the human musculature

  • Wilhelms J, VanGelder A. Anatomically based modeling. In: Int. conference on computer graphics and interactive...
  • J. Wilhelms

    Animals with anatomy

    IEEE Computer Graphics and Applications

    (1997)
  • Nedel LP, Thalmann D. Modeling and deformation of the human body using an anatomically-based approach. In: Proceedings...
  • Kähler K., Haber J., Seidel H.P.. Geometry-based muscle modeling for facial animation. In: Proceedings Graphics...
  • Albrecht I, Haber J, Seidel HP. Construction and animation of anatomically based human hand models. In: Proc....
  • Cited by (8)

    • Dynamic measurement of surface strain distribution on the foot during walking

      2017, Journal of the Mechanical Behavior of Biomedical Materials
      Citation Excerpt :

      3D DIC has recently been used to measure strain in biological tissues (e.g., Zhang and Arola, 2004; Moerman et al., 2009; Dickinson et al., 2011; Luyckx et al., 2014). However, although efforts have been made to estimate or simulate deformation and surface strain distribution of the foot (e.g., Marreiros, 2010; Fontanella et al., 2014; Tang and Hui, 2007), no attempts have been made to capture foot deformation and surface strain distribution during walking. Five adult male participants (173.1±16.9 cm, 67.8±12.8 kg) were asked to walk at a self-selected speed across a glass platform (0.2×0.38×0.01 m) set in a 5-m wooden walkway.

    • Footwear bio-modelling: An industrial approach

      2013, CAD Computer Aided Design
      Citation Excerpt :

      This model incorporates the salient features of muscle and tendon, specifies the musculoskeletal geometry and musculotendon parameters of eighteen musculotendon actuators, and defines the active isometric moment about the knee and ankle joints. In [34], the effect of the human foot’s tendons on the deformation of the foot surface was analysed. This stresses the importance that modelling has, not only of muscles and bones, but tendons as well as there are also certain forces that have effects on the system’s modelling and simulation, affecting the final foot surface deformation.

    • Human fringe skeleton extraction by an improved Hopfield neural network with direction features

      2012, Neurocomputing
      Citation Excerpt :

      Based on the shape segmentation, the vivid human models can be generated by modeling and deformation methods using the extracted skeleton. Tang and Hui [18] propose an algorithm for modeling human foot tendons and determines the influence on the skin deformation using the extracted skeleton. There exist a large number of research efforts on extracting high-quality skeleton.

    • Human foot modeling towards footwear design

      2011, CAD Computer Aided Design
      Citation Excerpt :

      Approaches for human foot modeling and animation have been investigated for many years [1–6].

    • Simulating tendon motion with axial mass-spring system

      2009, Computers and Graphics (Pergamon)
    View all citing articles on Scopus
    View full text