Manik, Fionaldy Andrianus (2024) Deteksi Ketepatan Gerak Pada Olahraga Angkat Beban. Other thesis, Institut Teknologi Sepuluh Nopember.
![]() |
Text
5024201038_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (51MB) | Request a copy |
Abstract
Penelitian ini berfokus pada pengembangan dan evaluasi sebuah sistem yang menggunakan MediaPipe dan LSTM (Long Short-Term Memory) untuk analisis gerakan dalam latihan squat dan deadlift. Tujuan utama dari penelitian ini adalah untuk mengevaluasi efektivitas dan akurasi MediaPipe, sebuah solusi berbasis machine learning untuk estimasi pose, dalam mengidentifikasi postur yang tepat selama latihan squat dan deadlift. Selanjutnya, LSTM digunakan untuk menganalisis data waktu-nyata dari gerakan tersebut, memungkinkan penilaian berkelanjutan terhadap kualitas pelaksanaan latihan. Hasil penelitian ini menunjukkan potensi penerapan kombinasi MediaPipe dan LSTM dalam membantu atlet dan individu untuk meningkatkan teknik latihan angkat beban mereka, serta dalam mencegah risiko cedera akibat teknik yang salah. Kedua alat ini, ketika digunakan bersama, memberikan wawasan yang berharga dalam aspek koreksi gerakan dan pelatihan olahraga, menandakan langkah signifikan dalam teknologi pelatihan kebugaran dan rehabilitasi.
========================================================================================================================
This research focuses on developing and evaluating a system that uses MediaPipe and LSTM (Long Short-Term Memory) for movement analysis in squat and deadlift training. The primary goal of this study was to evaluate the effectiveness and accuracy of MediaPipe, a machine learning-based solution for pose estimation, in identifying appropriate posture during squat and deadlift exercises. Next, LSTM is used to analyze real-time data of the movement, allowing continuous assessment of the quality of exercise execution. The results of this study show the potential application of the combination of MediaPipe and LSTM in helping athletes and individuals to improve their weight lifting training techniques, as well as in preventing the risk of injury due to incorrect technique. These two tools, when used together, provide valuable insight into aspects of movement correction and sports training, marking a significant step in fitness training and rehabilitation technology.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | Squat, Deadlift, mediapipe, LSTM |
Subjects: | Q Science > QA Mathematics > QA76.6 Computer programming. Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) Q Science > QA Mathematics > QA76.9.I52 Information visualization |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis |
Depositing User: | Fionaldy Andrianus Manik |
Date Deposited: | 02 Aug 2024 02:28 |
Last Modified: | 02 Aug 2024 02:28 |
URI: | http://repository.its.ac.id/id/eprint/108737 |
Actions (login required)
![]() |
View Item |