Background: The evaluation of Human Movements based on Taekwondo poomsae (movement patterns) is inherently subjective, leading to concerns about bias and inconsistency in scoring. This emphasizes the need for objective and reliable scoring systems leveraging artificial intelligence technologies. Objectives: This study seeks to enhance the accuracy and fairness of Taekwondo poomsae scoring through the application of camera-based pose estimation and advanced neural network models. Design: A comparative analysis was conducted to evaluate the performance of machine learning models on a large-scale Taekwondo poomsae dataset. Methods: The analysis utilized a dataset comprising 902,306 labeled frames from 48 participants performing 62 distinct poomsae movements. Five models—LSTM, GRU, Simple RNN, Random Forest, and XGBoost—were evaluated using performance metrics, including accuracy, precision, recall, F1- score, and log loss. Results: The LSTM model outperformed all others, achieving an accuracy, precision, recall, and F1-score of 0.81, alongside the lowest log loss value of 0.55. The GRU model demonstrated comparable performance, while traditional models such as Random Forest and XGBoost were less effective in capturing the temporal and spatial patterns of poomsae movements. Conclusion: The LSTM model exhibited superior capability in modeling the temporal and spatial complexities inherent in Taekwondo poomsae, establishing a robust foundation for the development of objective, scalable, and reliable poomsae evaluation systems.

PURPOSES : This study develops a model that can estimate travel speed of each movement flow using deep-learning-based probe vehicles at urban intersections. METHODS : Current technologies cannot determine average travel speeds for all vehicles passing through a specific real-world area under obseravation. A virtual simulation environment was established to collect information on all vehicles. A model estimate turning speeds was developed by deep learning using probe vehicles sampled during information processing time. The speed estimation model was divided into straight and left-turn models, developed as fully-offset, non-offset, and integrated models. RESULTS : For fully-offset models, speed estimation for both straight and left-turn models achieved MAPE within 10%. For non-offset models, straight models using data drawn from four or more probe vehicles achieved a MAPE of less than 15%. The MAPE for left turns was approximately 20%. CONCLUSIONS : Using probe-vehicle data(PVD), a deep learning model was developed to estimate speeds each movement flow. This, confirmed the viability of real-time signal control information processing using a small number of probe vehicles.

Two experiments were conducted to examine the effects of visual input enhancement (VIE) of target forms and deliberate attention on grammar learning and reading comprehension of Korean high school students. In Experiment 1, eighty-eight students read one of the three experimental texts: (i) BT (baseline text), (ii) VIE (BT with the target forms visually enhanced), and (iii) VIE-Attention (VIE with explicit instruction asking students to pay deliberate attention to both the target forms and reading comprehension). After reading, the students responded to grammar and reading comprehension tests. The results showed that only VIE-Attention promoted grammar learning, while both the VIE and VIE-Attention significantly impaired reading comprehension. In Experiment 2, an eye tracker was used in order to further probe the effects of VIE and deliberate attention. The results revealed that the VIE and VIEAttention groups fixated longer and more often than those in the BT group and that the VIE and VIE-Attention groups performed better in the grammar test and poorer on the reading comprehension test than the BT group. The present study makes significant contributions to the VIE literature since it provides the first eye movement data elucidating the effects of VIE.

The purpose of this study was to show the efficient aspect and the proper time of feedback for the ball movement teaching when the rhythmic gymnastics exercises are practiced in the class room situation, concerning that the method and the time of feedback

본 연구는 기호를 활용한 인지학습기반 표현움직임 프로그램의 현장 적용성을 탐색하는 것에 그 목적을 두고 있다. 먼저 인지학습 기반의 표현움직임 프로그램 구성에 활용된 Motif Writing 기호를 소개하여 전반적인 이해를 돕고자 하였다. 그리고 남 여 구분 없이 4-5세 유아 실험집단(9명)에게 이 연구를 위해 개발된 움직임교육 프로그램을 8주 동안 적용하는 현장 실험을 실시하고, 인지수업(영어), 신체활동수업 (발레), 스포츠 활동수업(수영) 각 9명씩 총 36명으로 이루어진 집단과 비교하여 다중지능에 있어서의 상대적인 변화를 살펴보았다. 분석 결과, 대인관계, 신체지능을 제외한 언어, 논리-수학, 공간, 음악, 개인 이해지능에서 비교집단보다 상대적으로 유의미한 향상이 나타났다. 이러한 결과는 기호를 활용한 인지학습기반 표현움직임 프로그램이 기호를 읽고 해석하고 자신의 생각을 움직임으로 표현하는 과정을 통해 다중지능의 전반적인 향상에 긍정적으로 작용한 것으로 보인다. 이는 본 프로그램이 단순히 신체적 발달뿐만 아니라 인지적, 정서적 발달에 이르는 전인적 교육 프로그램의 발전적 대안으로 활용될 수 있음을 확인함으로써, Motif Writing 기호를 활용한 표현 움직임 프로그램의 현장 적용 가능성을 시사해 주는 것이다.

Kernodle과 Carlton(1992)은 다자유도 동작의 학습에서 동영상 피드백(VF)의 효율을 최대화하기 위해 전이적 정보(TI)와 같이 결합하여 사용되어야 한다고 주장하였다. 그러나 그들의 연구에서 발견된 TI+VF의 학습효과는 동영상 피드백과는 관계없이 전이적 정보에 의한 효과였다고 주장될 수 있었다. 본 연구는 동영상 피드백의 유용성과 학습효율을 조사하기 위해 TI+VF 조건의 학습효과가 무엇으로 기인하였는지 검증하였다. 36명의 피험자들은 왼손으로 공 던지기를 하루 75회씩 2일간 연습하였다. 연습 동안 3회 시행마다 제공된 피드백의 유형에 따라 피험자들은 세 집단에 무작위로 배치되었다, (1)KR, (2)TI, 그리고 (3)TI+VF. 연습단계 24시간 후 실시된 파지검사 결과, 동작 폼의 경우 TI+VF 집단은 가장 우수한 학습효과를 나타냈다. TI 집단 또한 KR 집단보다 더 우수한 학습효과를 나타냈다. 수행결과를 나타내는 던진 거리의 경우 TI+VF 집단은 TI와 KR 집단보다 더 우수한 파지수행을 나타냈지만 그 차이는 유의함에 미치지 못했다. 이 결과는 전이적 정보가 다자유도 동작의 학습을 위해 효과적인 피드백 정보이지만 동영상 피드백이 보강될 때 학습효과는 유의하게 더 커진다는 것을 나타낸다. TI+VF는 동작협응과 수행결과의 학습에 각기 다른 속도로 영향을 미치는 것처럼 보인다. TI+VF의 제공은 동작협응에 대한 학습을 빠르게 향상시키는 반면, 수행결과에는 비교적 느리게 영향을 미친다.

The purpose of this research is to find the effect of motor learning if quantitative knowledge of results and qualitative knowledge of results are presented according to task natures. As the experimental method, I execute a kind of linear positioning task adopting 36 college students as objects of this study. After measuring movement time and distant error using a touch pad adhered a circuit and a programmed personal computer system, I calculate the mean of distant error and standard deviation. I come to get these results as followings to perform factorial ANOVA with repeated measures providing a basis for this computation. The task natures and forms of knowledge of results have not influences on the effect of motor learning. The performance effect according to task natures makes a meaningful difference, that is students show high accuracy of motion when they have enough time to perform the task voluntarily than under the condition that they have to do the task as soon as possible.