Visual Measurement of the Racket Trajectory in Spinning Ball Striking for Table Tennis Player

Chen, Guodong; Xu, De; Fang, Zaojun; Jiang, Zemin; Tan, Min

doi:10.1109/tim.2013.2265471

Cited by 25 publications

(14 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 3 shows the experimental results after using various numbers of cameras (3)(4)(5) in the multi-camera system. In this experiment, we randomly chose 25 poses for the multi-camera system, which were distributed in the half sphere towards the original centre (0,0,0) with a distance of 1,500, and then executed the estimation 200 times to output the average.…”

Section: Simulation Experiments On Other Metricsmentioning

confidence: 99%

“…Pose estimation from a referenced rigid structure is one of the most basic and important problems in robotics vision and computer graphics. It can be used to obtain the 6DOF (degrees of freedom) of the cameras needed for further implementation, e.g., in helping robots to locate targets in referenced coordinates [1, 2], in calculating coordinates in images of virtual 3D objects to synthesize augmented reality scenes [3], in locating flying balls for robots [4], and in calibrating the camera-laser sensor system [5].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Robust and Accurate Multiple-Camera Pose Estimation toward Robotic Applications

Liu

Xiong

2014

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

Pose estimation methods in robotics applications frequently suffer from inaccuracy due to a lack of correspondence and real-time constraints, and instability from a wide range of viewpoints, etc. In this paper, we present a novel approach for estimating the poses of all the cameras in a multi-camera system in which each camera is placed rigidly using only a few coplanar points simultaneously. Instead of solving the orientation and translation for the multi-camera system from the overlapping point correspondences among all the cameras directly, we employ homography, which can map image points with 3D coplanar-referenced points.In our method, we first establish the corresponding relations between each camera by their Euclidean geometries and optimize the homographies of the cameras; then, we solve the orientation and translation for the optimal homographies. The results from simulations and real case experiments show that our approach is accurate and robust for implementation in robotics applications. Finally, a practical implementation in a ping-pong robot is described in order to confirm the validity of our approach.

show abstract

Section: Simulation Experiments On Other Metricsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Robust and Accurate Multiple-Camera Pose Estimation toward Robotic Applications

Liu

Xiong

2014

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

show abstract

“…In [18], we propose an iterative algorithm to calculate the maximum pressure points and use them as the contact points to fit the trajectory. In [23]- [25], trajectories are solved and fitted through a large number of iterations.…”

Section: Introductionmentioning

confidence: 99%

A Wireless Trajectory Measurement System for Unicondylar Knee Arthroplasty

Wang

Zhou

et al. 2020

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

During unicondylar knee arthroplasty surgery, the tibial prosthesis is fixed and the angle of distal femoral prosthesis relative to the prosthetic gasket is changed to generate the force line trajectory, which is important for doctors to judge the appropriate position of the implants. In this paper, we propose a wireless force line trajectory measurement system for unicondylar knee arthroplasty, which includes the pressure sensors array, the control and processing data circuits, the radio frequency transmitter, and the data recorder. Data from twenty sensors is processed and transmitted wirelessly, and the force line trajectory is displayed on the screen in real-time. The least squares fitting algorithm is proposed, with which we obtain the pressure distribution, the maximum pressure points and the force line trajectory. We adopt field programmable gate array to implement the algorithm with lower-upper decomposition to fit the trajectory faster and transmit less data to save power. The average current consumption of the system is 1.37mA in the operating mode. Dimensions of the data transmission module are length 70mm, width 35mm and height 4mm. The total uncertainty of the system measurement force line trajectory is 0.92mm. Experimental results show that the system meets the requirement of real-time measurement of the force line trajectory during the surgery.

show abstract

“…With single view tting techniques, the 3D racket position was estimated with a spatial accuracy of 1.9 ± 0.14 mm. Chen et al [4] established a high-speed monocular vision system to track a table tennis racket labeled with some special marker lines in the form of a black rectangle in the middle and a white line parallel to one of the black lines. They can be extracted into several corners as feature points and the pose is computed based on perspective-n-point and orthogonal iteration algorithms.…”

Section: Introductionmentioning

confidence: 99%

“…The success rates for detection and classication did reach 95.7% and 96.7%, respectively. Zhang et al [21] fused inertial measurement unit (IMU) data with the method [4] based on an extended Kalman Filter for obtaining an accurate and robust racket pose. The racket position was computed from cameras and its orientation was estimated from both cameras and IMU resulting in an average angle error of 1.1 • .…”

Section: Introductionmentioning

confidence: 99%

Real-time 6D Racket Pose Estimation and Classification for Table Tennis Robots

Gao

2019

IJRC

View full text Add to dashboard Cite

For table tennis robots, it is a significant challenge to understand the opponent's movements and return the ball accordingly with high performance. One has to cope with various ball speeds and spins resulting from different stroke types. In this paper, we propose a real-time 6D racket pose detection method and classify racket movements into five stroke categories with a neural network. By using two monocular cameras, we can extract the racket's contours and choose some special points as feature points in image coordinates. With the 3D geometrical information of a racket, a wide baseline stereo matching method is proposed to find the corresponding feature points and compute the 3D position and orientation of the racket by triangulation and plane fitting. Then, a Kalman filter is adopted to track the racket pose, and a multilayer perceptron (MLP) neural network is used to classify the pose movements. We conduct two experiments to evaluate the accuracy of racket pose detection and classification, in which the average error in position and orientation is around 7.8 mm and 7.2 by comparing with the ground truth from a KUKA robot. The classification accuracy is 98%, the same as the human pose estimation method with Convolutional Pose Machines (CPMs).

show abstract

Visual Measurement of the Racket Trajectory in Spinning Ball Striking for Table Tennis Player

Cited by 25 publications

References 13 publications

Robust and Accurate Multiple-Camera Pose Estimation toward Robotic Applications

Robust and Accurate Multiple-Camera Pose Estimation toward Robotic Applications

A Wireless Trajectory Measurement System for Unicondylar Knee Arthroplasty

Real-time 6D Racket Pose Estimation and Classification for Table Tennis Robots

Contact Info

Product

Resources

About