Detection of Taekwondo Kicks Using RGB-D Sensors

Taekwondo has evolved from a traditional martial art into an official Olympic sport. This study introduces a novel action recognition model tailored for Taekwondo unit actions, utilizing joint-motion data acquired via wearable inertial measurement unit (IMU) sensors. The utilization of IMU sensor-measured motion data facilitates the capture of the intricate and rapid movements characteristic of Taekwondo techniques. The model, underpinned by a conventional convolutional neural network (CNN)-based image classification framework, synthesizes action images to represent individual Taekwondo unit actions. These action images are generated by mapping joint-motion profiles onto the RGB color space, thus encapsulating the motion dynamics of a single unit action within a solitary image. To further refine the representation of rapid movements within these images, a time-warping technique was applied, adjusting motion profiles in relation to the velocity of the action. The effectiveness of the proposed model was assessed using a dataset compiled from 40 Taekwondo experts, yielding remarkable outcomes: an accuracy of 0.998, a precision of 0.983, a recall of 0.982, and an F1 score of 0.982. These results underscore this time-warping technique’s contribution to enhancing feature representation, as well as the proposed method’s scalability and effectiveness in recognizing Taekwondo unit actions.

Section: Introductionmentioning

confidence: 99%

Action Recognition of Taekwondo Unit Actions Using Action Images Constructed with Time-Warped Motion Profiles

Lim,

Luo,

Lee

et al. 2024

“…Castaneda et al, developed a PSS that attaches an inertial sensor to the body of a Taekwondo protector to detect the typical kicking techniques used in Taekwondo [ 4 ]. J. de Goma et al, used an RGB-D sensor and skeletal data from Kinect to determine Taekwondo behavior with a preprocessing-focused approach [ 5 ]. Dharmayanti et al, analyzed the kinematic characteristics of a Taekwondo punch technique using a motion analysis camera and inertial sensor [ 6 ].…”

Section: Introductionmentioning

confidence: 99%

Development of an Inertial Sensor Module for Categorizing Anomalous Kicks in Taekwondo and Monitoring the Level of Impact

Jang

Lee

et al. 2022

In this study, an inertial measurement unit (IMU) sensor module and software algorithm were developed to identify anomalous kicks that should not be given scores in Taekwondo competitions. The IMU sensor module was manufactured with dimensions of 3 cm × 3 cm × 1.5 cm and consists of a high-g sensor for high acceleration measurement, a 9-DOF sensor, and a Wi-Fi module for wireless communication. In the experiment, anomalous kicks and normal kicks were collected by the IMU sensor module, and an AI model was trained. The anomalous kick determination accuracy of the trained AI model was found to be 97.5%. In addition, in order to check whether the strength of a blow can be distinguished using the IMU sensor module, an impact test was performed with a pendulum under the same test conditions as the impact sensor installed in the impact test setup, and the correlation coefficient was 0.99. This study is expected to contribute to improving scoring reliability by suggesting the possibility of discriminating anomalous kicks, which were difficult to judge in Taekwondo competitions, through the analysis of Taekwondo kicks using inertial data and impulses.

“…Recent advances in artificial intelligence technology and vision sensors have promoted vision-based action recognition for various applications, such as education [ 2 ], entertainment [ 3 , 4 ], and sports [ 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 ]. Various studies have proposed novel algorithms [ 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 ] or established datasets [ 1 , 24 , 25 , 26 , 27 ] for vision-based action recognition.…”

Section: Introductionmentioning

confidence: 99%

TUHAD: Taekwondo Unit Technique Human Action Dataset with Key Frame-Based CNN Action Recognition

Lee

Jung

2020

In taekwondo, poomsae (i.e., form) competitions have no quantitative scoring standards, unlike gyeorugi (i.e., full-contact sparring) in the Olympics. Consequently, there are diverse fairness issues regarding poomsae evaluation, and the demand for quantitative evaluation tools is increasing. Action recognition is a promising approach, but the extreme and rapid actions of taekwondo complicate its application. This study established the Taekwondo Unit technique Human Action Dataset (TUHAD), which consists of multimodal image sequences of poomsae actions. TUHAD contains 1936 action samples of eight unit techniques performed by 10 experts and captured by two camera views. A key frame-based convolutional neural network architecture was developed for taekwondo action recognition, and its accuracy was validated for various input configurations. A correlation analysis of the input configuration and accuracy demonstrated that the proposed model achieved a recognition accuracy of up to 95.833% (lowest accuracy of 74.49%). This study contributes to the research and development of taekwondo action recognition.