Lejun Shen

Ph.D., Professor, Chengdu Sports University.

email: sljcool {At} sina.com ()

Publication:

Net-kill Opportunity Created by Smash in Badminton Doubles (IACSS 2023)

Abstract: Badminton is one of the most popular sports in the world. In this pa-per, we examined the assumption that smash in the badminton double discipline is tactically different from the single discipline and that there exists a powerful three-stroke sequence: cooperation of smash and net-kill (CoSN). Four evalua-tion criteria (direct scoring rate, create scoring rate, net-kill opportunity and awards) were proposed in this paper. Five smash parameters (height of impact-point, post-impact shuttlecock speed, distance from impact point to back boundary line, shuttle flight time and height of trajectory end-point) were measured and counted in a balls-into-bins model to investigate the link between scoring rate and smash parameters. We collected a dataset comprising 55,433 strokes from 46 world-class women doubles games. We found the most relevant smash parameter is shuttle flight time. The knowledge of low-cost-high-reward smash behaviour helps to improve training methods.

Lejun Shen, Yunlei Zhao, Yongming Chen, Ting Li, Ning Tang, Lu Ding and Jinwen Deng

Download: PDF (Full Text)

Measurement and Performance Evaluation of Lob Technique using Aerodynamic Model In Badminton Matches (IACSS 2019)

Abstract: In badminton matches, lob is a special technique and can be classified into two categories: defensive and offensive. These lobs are difficult to quantitatively measure, analyze, and evaluate. In this paper, we propose a new aerodynamic model to estimate the 3D trajectory from a single camera video and evaluate the performance of lobs. The experimental results show that this model is reliable. Offensive lobs are easily identified by the height of the trajectory. Good lobs are placed farther from the opponent than the bad lobs.

Lejun Shen, Hui Zhang, Min Zhu, Jia Zheng, Yawei Ren

Download: PDF (Full Text), Source Code

Reconstruction of 3D ball/shuttle position by two image points from a single view (IACSS 2017)

Abstract: Monocular 3D reconstruction is an important problem. We solve the problem of reconstructing a 3D ball or shuttle position from a single-view television video. The contextual constraint is vital in this paper, which is implemented by a confirming point. The confirming point represents all the contextual cues, such as human pose, stroke technique, and shadow on the ground. The confirming point also tells us where the 3D point is. The confirming point decision is made by a human operator. Thus, the proposed method is a mixture of computer vision and human intelligence. Moreover, we propose a new air-ball friction model. This model provides a more accurate result because the aerodynamic drag force cannot be ignored in ball game.

Lejun Shen, Qing Liu, Lin Li, Yawei Ren

Download: PDF (Full Text)

3D reconstruction of ball trajectory from a single camera in the ball game (ISCSS 2015)

Abstract: The 3D ball trajectory provides us quantitative technical or tactical information (e.g. the ball speed of serve). The 3D trajectory can be reconstructed by multiple camera system or single camera system. The single camera 3D reconstruction method is better than the multiple camera one, because it is convenient for the video from television. The existing monocular 3D reconstruction method suffers from the model-drifting problem. We solve this problem using a new cost function. Experimental result shows that our method is more accurate than classical method, because our cost function is a mixture of the physical model and the geometric model.

Lejun Shen, Qing Liu, Lin Li, Haipeng Yue

Download: PDF (Full Text)

Real-time Tracking of Multiple Objects by Linear Motion and Repulsive Motion (ACCV 2014)

Abstract: Successful multi-object tracking requires consistently maintaining object identities and real-time performance. This task becomes more challenging when objects are indistinguishable from one another. This paper presents a Bayesian framework for maintaining the identities of multiple objects. Our semi-independent joint motion model (SIMM) solves the coalescence and identity switching problem in real time. This joint motion model is a non-parametric mixture model that simultaneously captures linear motion and repulsive motion. Linear motion is a constant velocity model, while repulsive motion is described by a repulsive potential in MRF. By maintaining multimodality from multiple motion models, we can infer the appropriate motion model using image evidence and consequently avoid many identity switching errors. Moreover, we develop a new sampling method that does not suffer from the curse of dimensionality because of the availability of high-quality samples. Experimental results show that our approach can track numerous objects in real time and maintain identities under difficult situations.

Lejun Shen, Zhisheng You and Qing Liu

Download: PDF (Full Text) , Executable DEMO, Source Code