Motion Capture Method Research Articles

Finger Joint Angle Estimation from Electrical Contact Resistance

SUMMARYHand motion capture is an important topic for understanding the mechanism of the hand. General hand motion capture methods using cameras or bent sensors can capture the limited action in the restricted environments. This study proposes a novel hand motion capture method without restrictions of usage environment and user’ action. The proposed method is based on the electrical contact resistance between the wrist skin and the electrode to measure the deformation of the wrist which is related to the hand motion. We fabricate the contact resistance measurement circuit consists of three electrodes and a multiplexer. The output voltage corresponding to each contact resistance of the electrode is measured with the circuit by switching the working electrode temporally. We confirm that the output voltage is related to the wrist shape and changes according to the hand posture. Furthermore, the result suggests that the finger joint angles can be estimated from the output voltage due to the correlation between them.

Outdoor Markerless Motion Capture with Sparse Handheld Video Cameras.

We present a method for outdoor markerless motion capture with sparse handheld video cameras. In the simplest setting, it only involves two mobile phone cameras following the character. This setup can maximize the flexibilities of data capture and broaden the applications of motion capture. To solve the character pose under such challenge settings, we exploit the generative motion capture methods and propose a novel model-view consistency that considers both foreground and background in the tracking stage. The background is modeled as a deformable 2D grid, which allows us to compute the background-view consistency for sparse moving cameras. The 3D character pose is tracked with a global-local optimization through minimizing our consistency cost. A novel motion regularizer is also proposed in the optimization to constrain the solution pose space. The whole process of the proposed method is simple as frame by frame video segmentation is not required. Our method outperforms several alternative methods on various examples demonstrated in the paper.

Markerless Motion Capture for Entrance Guard Systems

In recent years, markerless motion capture is used to track human motions in a wide range of applications, from entertainment to surveillance. The markerless-based technology is more flexible and less cumbersome in comparison with marker-based optical systems. In this paper, we present a practical markerless motion capture method and make use of motion estimation for developing an entrance guard system. In the first step, a binary image is calculated to carry out background subtraction and region partition. Afterwards, we reconstruct an articulated skeleton model to measure the model-to-image similarity with the binary image in order to possess locations of the specific end joints. In addition, a decision tree is defined in advance to facilitate pose estimation. As a result, a valid or invalid command can be determined to control the door lock through the proposed entrance guard system. In our experiments, we demonstrate that our method is feasible to capture human motions and is applicable to the entrance guard system.

Model-based extended quaternion Kalman filter to inertial orientation tracking of arbitrary kinematic chains.

Inertial orientation tracking is still an area of active research, especially in the context of out-door, real-time, human motion capture. Existing systems either propose loosely coupled tracking approaches where each segment is considered independently, taking the resulting drawbacks into account, or tightly coupled solutions that are limited to a fixed chain with few segments. Such solutions have no flexibility to change the skeleton structure, are dedicated to a specific set of joints, and have high computational complexity. This paper describes the proposal of a new model-based extended quaternion Kalman filter that allows for estimation of orientation based on outputs from the inertial measurements unit sensors. The filter considers interdependencies resulting from the construction of the kinematic chain so that the orientation estimation is more accurate. The proposed solution is a universal filter that does not predetermine the degree of freedom at the connections between segments of the model. To validation the motion of 3-segments single link pendulum captured by optical motion capture system is used. The next step in the research will be to use this method for inertial motion capture with a human skeleton model.

EgoCap

Marker-based and marker-less optical skeletal motion-capture methods use an outside-in arrangement of cameras placed around a scene, with viewpoints converging on the center. They often create discomfort with marker suits, and their recording volume is severely restricted and often constrained to indoor scenes with controlled backgrounds. Alternative suit-based systems use several inertial measurement units or an exoskeleton to capture motion with an inside-in setup, i.e. without external sensors. This makes capture independent of a confined volume, but requires substantial, often constraining, and hard to set up body instrumentation. Therefore, we propose a new method for real-time, marker-less, and egocentric motion capture: estimating the full-body skeleton pose from a lightweight stereo pair of fisheye cameras attached to a helmet or virtual reality headset - an optical inside-in method, so to speak. This allows full-body motion capture in general indoor and outdoor scenes, including crowded scenes with many people nearby, which enables reconstruction in larger-scale activities. Our approach combines the strength of a new generative pose estimation framework for fisheye views with a ConvNet-based body-part detector trained on a large new dataset. It is particularly useful in virtual reality to freely roam and interact, while seeing the fully motion-captured virtual body.

Temporal–spatial reach parameters derived from inertial sensors: Comparison to 3D marker-based motion capture

Reaching is a well-practiced functional task crucial to daily living activities, and temporal–spatial measures of reaching reflect function for both adult and pediatric populations with upper-extremity motor impairments. Inertial sensors offer a mobile and inexpensive tool for clinical assessment of movement. This research outlines a method for measuring temporal–spatial reach parameters using inertial sensors, and validates these measures with traditional marker-based motion capture. 140 reaches from 10 adults, and 30 reaches from nine children aged 18–20 months, were recorded and analyzed using both inertial-sensor and motion-capture methods. Inertial sensors contained three-axis accelerometers, gyroscopes, and magnetometers. Gravitational offset of accelerometer data was measured when the sensor was at rest, and removed using sensor orientation measured at rest and throughout the reach. Velocity was calculated by numeric integration of acceleration, using a null-velocity assumption at reach start. Sensor drift was neglected given the 1–2s required for a reach. Temporal–spatial reach parameters were calculated independently for each data acquisition method. Reach path length and distance, peak velocity magnitude and timing, and acceleration at contact demonstrated consistent agreement between sensor- and motion-capture-based methods, for both adult and toddler reaches, as evaluated by intraclass correlation coefficients from 0.61 to 1.00. Taken together with actual difference between method measures, results indicate that these functional reach parameters may be reliably measured with inertial sensors.

Assessing trans-femoral residuum/socket interface coupling using 3D motion capture – effect of terrains and walking speeds

The aim of this study was to develop a new means to assess residuum/socket interface couplings at different speeds and walking over different terrains using motion capture methods and a conventional 6 degrees of freedom (6DoF) marker model.This dataset accompanies the following paper: Tang, Jing, Mcgrath, Michael, Hale, Nicholas, Jiang, Liudi, Bader, Dan, Moser, David, Laszczak, Piotr, Bradbury, Richard and Zahhedi, Saeed (2016) Assessing trans-femoral residuum/socket interface coupling using 3D motion capture – effect of terrains and walking speeds. Gait & Posture

Validation of the greater trochanter method with radiographic measurements of frontal plane hip joint centers and knee mechanical axis angles and two other hip joint center methods

Several motion capture methods exist for predicting hip joint centers (HJC). These methods include regression models, functional joints, and projections from greater trochanters. While regression and functional methods have been compared to imaging techniques, the TROCH method has not been previously validated. The purpose of this study was to compare frontal-plane HJCs and knee mechanical axis angles estimated using the greater trochanter method with a regression (Bell) and a functional method against those obtained using radiographs. Thirty-five participants underwent a long-standing anteroposterior radiograph, and performed static and functional motion capture trials. The Bell, functional, and trochanter HJCs were constructed to predict mechanical axes and compare HJC locations. One-way repeated measures ANOVAs were used to compare mechanical axes and HJC locations estimated by motion capture methods and measured using radiographs (p<0.05). All methods overestimated mechanical axes compared to radiographs (<2°), but were not different. Mediolateral HJC locations and inter-HJC widths were similar between methods; however, inter-HJC widths were underestimated (average 3.7%) compared to radiographs. The Bell HJC was more superior and anterior to both functional and trochanter methods. The trochanter HJC was more posterior to both methods. The Bell method outperformed the other methods in leg length predictions compared to radiographs. Although differences existed between methods, all frontal-plane HJC location differences were <1.7cm. This study validated the trochanter HJC prediction method mediolaterally and vertically (with small respective correction factors). Therefore, all HJC methods seem to be viable in predicting mechanical axes and frontal-plane HJC locations compared with radiographs.

Characterizing spatiotemporal information loss in sparse-sampling-based dynamic MRI for monitoring respiration-induced tumor motion in radiotherapy.

Sparse-sampling and reconstruction techniques represent an attractive strategy to achieve faster image acquisition speeds, while maintaining adequate spatial resolution and signal-to-noise ratio in rapid magnetic resonance imaging (MRI). The authors investigate the use of one such sequence, broad-use linear acquisition speed-up technique (k-t BLAST) in monitoring tumor motion for thoracic and abdominal radiotherapy and examine the potential trade-off between increased sparsification (to increase imaging speed) and the potential loss of "true" information due to greater reliance on a priori information. Lung tumor motion trajectories in the superior-inferior direction, previously recorded from ten lung cancer patients, were replayed using a motion phantom module driven by an MRI-compatible motion platform. Eppendorf test tubes filled with water which serve as fiducial markers were placed in the phantom. The modeled rigid and deformable motions were collected in a coronal image slice using balanced fast field echo in conjunction with k-t BLAST. Root mean square (RMS) error was used as a metric of spatial accuracy as measured trajectories were compared to input data. The loss of spatial information was characterized for progressively increasing acceleration factor from 1 to 16; the resultant sampling frequency was increased approximately from 2.5 to 19 Hz when the principal direction of the motion was set along frequency encoding direction. In addition to the phantom study, respiration-induced tumor motions were captured from two patients (kidney tumor and lung tumor) at 13 Hz over 49 s to demonstrate the impact of high speed motion monitoring over multiple breathing cycles. For each subject, the authors compared the tumor centroid trajectory as well as the deformable motion during free breathing. In the rigid and deformable phantom studies, the RMS error of target tracking at the acquisition speed of 19 Hz was approximately 0.3-0.4 mm, which was smaller than the reconstructed pixel resolution of 0.67 mm. In the patient study, the dynamic 2D MRI enabled the monitoring of cycle-to-cycle respiratory variability present in the tumor position. It was seen that the range of centroid motion as well as the area covered due to target motion during each individual respiratory cycle was underestimated compared to the entire motion range observed over multiple breathing cycles. The authors' initial results demonstrate that sparse-sampling- and reconstruction-based dynamic MRI can be used to achieve adequate image acquisition speeds without significant information loss for the task of radiotherapy guidance. Such monitoring can yield spatial and temporal information superior to conventional offline and online motion capture methods used in thoracic and abdominal radiotherapy.

A wireless accelerometer node for reliable and valid measurement of lumbar accelerations during treadmill running

This study investigated the reliability of a wireless accelerometer and its agreement with optical motion capture for the measurement of root mean square (RMS) acceleration during running. RMS acceleration provides a whole-body metric of movement mechanics and economy. Fifteen healthy college-age participants performed treadmill running for two 60-s trials at 2.22, 2.78, and 3.33 m/s and one trial of 150 s (five 30-s epochs) at 2.78 m/s. We assessed between-trial and within-trial reliability, and agreement in each axis between a trunk-mounted wireless accelerometer and a reflective marker on the accelerometer measured by optical motion capture. Intraclass correlations assessing between-trial repeatability were 0.89–0.97, depending on the axis, and intraclass correlations assessing within-trial repeatability were 0.99–1.00. Bland–Altman analyses assessing agreement indicated mean difference values between −0.03 and 0.03 g, depending on the axis. Anterio-posterior acceleration had the greatest limits of agreement (LOA) (±0.12 g) and vertical acceleration had the smallest LOA (±0.03 g). For measuring RMS acceleration of the trunk, this wireless accelerometer node provides repeatable and valid measurement compared with the standard laboratory method of optical motion capture.

Flexible CNT-array double helices Strain Sensor with high stretchability for Motion Capture.

Motion capture is attracting more and more attention due to its potential wide applications in various fields. However, traditional methods for motion capture still have weakness such as high cost and space consuming. Based on these considerations, a flexible, highly stretchable strain sensor with high gauge factor for motion capture is fabricated with carbon nanotube (CNT) array double helices as the main building block. Ascribed to the unique flexible double helical CNT-array matrix, the strain sensor is able to measure strain up to 410%, with low hysteresis. Moreover, a demonstration of using this strain sensor for capture hand motion and to control a mechanical hand in real time is also achieved. A model based on finite difference method is also made to help understand the mechanism of the strain sensors. Our work demonstrates that strain sensors can measure very large strain while maintaining high sensitivity, and the motion capture based on this strain sensor is expected to be less expensive, more convenient and accessible.

An instrumented spatial linkage for measuring knee joint kinematics

BackgroundIn this study, the design and development of a highly accurate instrumented spatial linkage (ISL) for kinematic analysis of the ovine stifle joint is described. The ovine knee is a promising biomechanical model of the human knee joint. MethodsThe ISL consists of six digital rotational encoders providing six degrees of freedom (6-DOF) to its motion. The ISL makes use of the complete and parametrically continuous (CPC) kinematic modeling method to describe the kinematic relationship between encoder readings and the relative positions and orientation of its two ends. The CPC method is useful when calibrating the ISL, because a small change in parameters corresponds to a small change in calculated positions and orientations and thus a smaller optimization error, compared to other kinematic models. The ISL is attached rigidly to the femur and the tibia for motion capture, and the CPC kinematic model is then employed to transform the angle sensor readings to relative motion of the two ends of the linkage, and thereby, the stifle joint motion. ResultsThe positional accuracy for ISL after calibration and optimization was 0.3±0.2mm (mean +/− standard deviation). The ISL was also evaluated dynamically to ensure that accurate results were maintained, and achieved an accuracy of 0.1mm. ConclusionsCompared to the traditional motion capture methods, this system provides increased accuracy, reduced processing time, and ease of use. Future work will be on the application of the ISL to the ovine gait and determination of in vivo joint motions and tissue loads. Clinical relevanceAccurate measurement of knee joint kinematics is essential in understanding injury mechanisms and development of potential preventive or treatment strategies.

Leveraging Two Kinect Sensors for Accurate Full-Body Motion Capture.

Accurate motion capture plays an important role in sports analysis, the medical field and virtual reality. Current methods for motion capture often suffer from occlusions, which limits the accuracy of their pose estimation. In this paper, we propose a complete system to measure the pose parameters of the human body accurately. Different from previous monocular depth camera systems, we leverage two Kinect sensors to acquire more information about human movements, which ensures that we can still get an accurate estimation even when significant occlusion occurs. Because human motion is temporally constant, we adopt a learning analysis to mine the temporal information across the posture variations. Using this information, we estimate human pose parameters accurately, regardless of rapid movement. Our experimental results show that our system can perform an accurate pose estimation of the human body with the constraint of information from the temporal domain.

手首形状に応じた電気接触抵抗からの指角度推定

Hand motion capture is an important topic for understanding the mechanism of the hand. General hand motion capture methods using cameras or bending sensors can capture the limited action in the restricted environments. This study proposes a novel hand motion capture method without restrictions of usage environment and user's action. The proposed method is based on the electrical contact resistance between the wrist skin and the electrode to measure the deformation of the wrist which is related to the hand motion. We fabricate the contact resistance measurement circuit consists of 3 electrodes and a multiplexer. The output voltage corresponding to each contact resistance of the electrode is measured with the circuit by switching the active electrode temporally. We confirm that the output voltage is related to the wrist shape and changes according to the hand posture. Furthermore, the result suggests that the finger joint angles can be estimated from the output voltage due to the correlation between them.

Capturing Human Motion in Natural Environments

The problem of capturing human motion in a natural environment is discussed from the perspective of needs, significance, scenarios, and technical challenges. The technologies that can be potentially used to capture human motion and activity in a natural environment are briefly discussed with an emphasis on computer vision-based markerless motion capture technology. Three representative markerless motion capture methods for capturing human motion from video imagery are implemented in this paper. The synthetic data generated by modeling and simulation and the data collected in laboratory environments are used for the training and validation. The initial results of three methods are presented and analyzed, and the advantages and disadvantages of each are discussed and compared.

Leveraging depth cameras and wearable pressure sensors for full-body kinematics and dynamics capture

We present a new method for full-body motion capture that uses input data captured by three depth cameras and a pair of pressure-sensing shoes. Our system is appealing because it is low-cost, non-intrusive and fully automatic, and can accurately reconstruct both full-body kinematics and dynamics data. We first introduce a novel tracking process that automatically reconstructs 3D skeletal poses using input data captured by three Kinect cameras and wearable pressure sensors. We formulate the problem in an optimization framework and incrementally update 3D skeletal poses with observed depth data and pressure data via iterative linear solvers. The system is highly accurate because we integrate depth data from multiple depth cameras, foot pressure data, detailed full-body geometry, and environmental contact constraints into a unified framework. In addition, we develop an efficient physics-based motion reconstruction algorithm for solving internal joint torques and contact forces in the quadratic programming framework. During reconstruction, we leverage Newtonian physics, friction cone constraints, contact pressure information, and 3D kinematic poses obtained from the kinematic tracking process to reconstruct full-body dynamics data. We demonstrate the power of our approach by capturing a wide range of human movements and achieve state-of-the-art accuracy in our comparison against alternative systems.

The Effect of Lunate Morphology on the 3-Dimensional Kinematics of the Carpus

To assess carpal kinematics in various ranges of motion in 3 dimensions with respect to lunate morphology. Eight cadaveric wrists (4 type I lunates, 4 type II lunates) were mounted into a customized platform that allowed controlled motion with 6 degrees of freedom. The wrists were moved through flexion-extension (15°-15°) and radioulnar deviation (RUD; 20°-20°). The relative motion of the radius, carpus, and third metacarpal were recorded using optical motion capture methods. Clear patterns of carpal motion were identified. Significantly greater motion occurred at the radiocarpal joint during flexion-extension of type I wrist than a type II wrist. The relative contributions of the midcarpal and radiocarpal articulations to movement of the wrist differed between the radial, the central, and the ulnar columns. During wrist flexion and extension, these contributions were determined by the lunate morphology, whereas during RUD, they were determined by the direction of wrist motion. The midcarpal articulations were relatively restricted during flexion and extension of a type II wrist. However, during RUD, the midcarpal joint of the central column became the dominant articulation. This study describes the effect of lunate morphology on 3-dimensional carpal kinematics during wrist flexion and extension. Despite the limited size of the motion arcs tested, the results represent an advance on the current understanding of this topic. Differences in carpal kinematics may explain the effect of lunate morphology on pathological changes within the carpus. Differences in carpal kinematics due to lunate morphology may have implications for the management of certain wrist conditions.

Research on Intangible Cultural Heritage Based on Motion Capture

The application of digital technology on the protection of intangible cultural heritage is a major topic of research in recent years. The motion capture technology of protection will gradually replace the traditional recording methods such as texts, pictures and videos. It is valuable to build a high-fidelity, high-modular and low-cost digital platform for choreographic data collection and extended application. This paper studies the intangible cultural heritage of Quanzhou breast-clapping dance, one of the most famous choreographic intangible cultural heritages from China with standard optical motion capture method. The data are acquiring and processing after the dance motion capture, we binds the motion data and three-dimensional model using Motion Builder and build digital demonstration platform base on an OGRE engine to display the movements. The viewer can view at any angle and distance. The system can be easily applied in motion intangible cultural heritages protection project. Furthermore, the system can be provided versatile motion data for additional use.

Research on Computer Modeling of Human Movement

Human body is a complex life forms.A lot of complexity means that affect motor function of factor structure. Human movement is a dynamic feedback process control, need to be coordinated through the visual, vestibular, proprioceptive feedback, not only by the physical laws of gravity, inertia force of domination, but also by the emotional impact of individual factors such as hobbies. Computer modeling of human actions including key frame method, kinematics, dynamics, motion capture method and process control methods. These methods have their own advantages, there are also their limitations. The different methods to combine the use of their complementary strengths will be the focus of future research and development direction.

Cable motion capture and analysis based on optical tracking system

SUMMARYAs a fatigable part of industrial products, cable is a valuable research topic to predict the product lifetime. Since the cable fatigue is related to the bending radius in motion, this paper presents a cable motion capture and bending radius calculation method based on optical tracking system. In particular, a marker sorting algorithm is developed for further spline interpolation in cable motion tracking. The cable bending radius calculation precision is analysed as well. The Results show that the proposed method can successfully track cable motion with an acceptable error for cable lifetime prediction.