Pengembangan Model Long Short-Term Memory Berbasis MediaPipe Pose untuk Klasifikasi dan Penilaian Gerakan Push-Up
Abstract
The counting and assessment of push-up movements are often inaccurate and subjective because they are done manually. Existing automated approaches generally only consider the top and bottom positions. This makes the assessment incomplete, so that imperfect movements can be counted as correct. This study aims to develop a Long Short-Term Memory (LSTM) model using MediaPipe Pose landmark extraction to detect and classify push-up movements based on the overall movement pattern. Data was collected in the form of videos validated by experts and processed into 3,600 video data sets with 10 movement classes. The video data was extracted to produce normalized keypoint coordinates and joint angles, and padding was added to equalize the data length. The model was trained using the K-Fold Cross Validation method with eight different architectures. The results showed that the best model achieved an average testing accuracy of 92.35% and an F1-Score of 92.53%. These findings indicate that the combination of MediaPipe Pose and LSTM can effectively recognize and classify push-up movements.
Keywords: MediaPipe Pose;Llong short-term memory; Classification; Push-up; Time series
Abstrak
Penghitungan dan penilaian gerakan push-up sering tidak akurat serta bersifat subjektif akibat dilakukan secara manual. Pendekatan otomatis yang ada umumnya hanya mempertimbangkan posisi atas dan bawah. Hal ini membuat penilaian tidak secara keseluruhan sehingga gerakan yang tidak sempurna dapat terhitung benar. Penelitian ini bertujuan mengembangkan model Long Short-Term Memory (LSTM) dengan memanfaatkan ektraksi landmark MediaPipe Pose untuk mendeteksi dan mengklasifikasi gerakan push-up berdasarkan keseluruhan pola gerakan. Data dikumpulkan dalam bentuk video yang divalidasi oleh ahli dan diolah hingga mejadi 3.600 data video dengan 10 kelas gerakan. Data video diektraksi hingga menghasilkan koordinat keypoint dan sudut sendi yang dinormalisasi, serta ditambahkan padding untuk menyamakan panjang data. Model dilatih dengan metode K-Fold Cross Validation dengan delapan arsitektur berbeda. Hasil penelitian menunjukkan performa model terbaik memperoleh rata-rata akurasi testing 92,35% dan F1-Score 92,53%. Temuan ini menunjukkan kombinasi MediaPipe Pose dan LSTM dapat dengan baik mengenali dan mengklasifikasi gerakan push-up.
Keywords
References
S. F. M. Chastin et al., “Effects of Regular Physical Activity on the Immune System, Vaccination and Risk of Community-Acquired Infectious Disease in the General Population: Systematic Review and Meta-Analysis,” Sports Medicine, vol. 51, no. 8, pp. 1673–1686, 2021, doi: 10.1007/s40279-021-01466-1.
F. M. Lulie, “Effects Of Calisthenics Exercise On Selected Health Related Physical Fitnes Qualities In The Case Of Tagel Secondary And Preparatory School Male Student”, Tesis, Department of Sport Science, Bahir Dar University, Ethiopia, 2024.
C. H. J. K. Otarsky, B. R. K. C. Hristensen, and J. A. S. M. Iller, “Effect Of Progressive Calisthenic Push-Up Training On Muscle Strength And Thickness,” J Strength Cond Res, vol. 32, no. 3, pp. 651–659, 2018, doi: 10.1519/JSC.0000000000002345.
I. W. Artanayasa, K. C. A. Kusuma, and K. U. Ariawan, “Need analysis of digital technology-based push up test: Instrument of arm muscle strength,” Journal Sport Area, vol. 7, no. 3, pp. 361–368, Dec. 2022, doi: 10.25299/sportarea.2022.vol7(3).10641.
I. W. Artanayasa, K. C. A. Kusuma, and K. U. Ariawan, “Push-Up Counter (PUC) as an instrument of arm muscle strength: Validity and reliability,” Journal Sport Area, vol. 8, no. 3, pp. 350–359, Sep. 2023, doi: 10.25299/sportarea.2023.vol8(3).13245.
R. Mardela, R. Irawan, A. Agung, N. L. Putra, Y. Marlina, and R. Ependi, “Development Of A Digital-Based Push Up And Sit Up Test Counter,” Jurnal Laman Olahraga Nusantara, pp. 287–298, 2023, doi: 10.31851/hon.v6i1.10723.
J. Zhang et al., “MobiPose: Real-time multi-person pose estimation on mobile devices,” in SenSys 2020 - Proceedings of the 2020 18th ACM Conference on Embedded Networked Sensor Systems, Association for Computing Machinery, Inc, Nov. 2020, pp. 136–149. doi: 10.1145/3384419.3430726.
S. P. Raj, S. V Raam, S. B. Gopi, and D. Kumar, “Real-Time Body Pose Estimation Using Opencv And Mediapipe,” International Journal of Creative Research Thoughts, vol. 12, pp. 2320–2882, 2024, doi: https://ijcrt.org/papers/IJCRT24A5046.pdf.
R. Arlin and R. Munir, “The Development of Push Up Counter Android Application with Computer Vision,” Bandung, 2021.
H.-J. Park, J.-W. Baek, and J.-H. Kim, Imagery based Parametric Classification of Correct and Incorrect Motion for Push-up Counter Using OpenPose. 2020. doi: 10.0/Linux-x86_64.
A. Y. Setiawan, I. G. M. Darmawiguna, and G. Pradnyana, “Sentiment Summarization Learning Evaluation Using LSTM (Long Short Term Memory) Algorithm,” European Alliance for Innovation (EAI), Singaraja, 1 – 17, 27 November, 2021, doi: 10.4108/eai.27-11-2021.2315533.
B. A. Wisesa, W. Andriyani, and B. Purnomosidi, “Usage of LSTM Method On Hand Gesture Recognition For Easy Learning of Sign Language Based On Desktop Via Webcam,” International Seminar on Research of Information Technology and Intelligent Systems (ISRITI), pp. 148–153, Mar. 2023, doi: 10.1109/ISRITI56927.2022.10053076.
G. B. Pratama Putra, P. H. Suputra, N. W. Marti, and K. F. Sugiantari, “Deteksi Jatuh Lansia Berbasis Landmark Sendi Pada Model LSTM,” Jurnal Informatika Polinema, vol. 11, no. 3, pp. 261–266, 2025, doi: 10.33795/jip.v11i3.6823.
I. M. P. Prayoga, I. G. Ayu, A. Diatri, P. H. Suputra, and P. August, “Pengenalan Kata Kolok Secara Real Time Menggunakan Mediapipe dan Algoritma Long Short-Term Memory ( LSTM ),” Science and Information Technology Journal, vol. 8, no. 2, pp. 178–187, 2025, doi: https://doi.org/10.31598/sintechjournal.v8i2.1945.
F. D. Tanugraha, H. Pratikno, M. Musayanah, and W. I. Kusumawati, “Pengenalan Gerakan Olahraga Berbasis (Long Short- Term Memory) Menggunakan Mediapipe,” Journal of Advances in Information and Industrial Technology, vol. 4, no. 1, pp. 37–45, May 2022, doi: 10.52435/jaiit.v4i1.182.
N. P. N. P. Dewi, N. K. Kertiasih, and N. L. D. Sintiari, “Modifikasi Fruit Fly Optimiziation Algorithm untuk Optimasi General Regression Neural Network pada Kasus Prediksi Time-Series,” Jurnal Nasional Pendidikan Teknik Informatika (JANAPATI), vol. 11, no. 3, pp. 192–204, 2022, doi: 10.23887/janapati.v11i3.54521.
N. K. T. A. Saputri, I. G. A. Gunadi, and I. M. G. Sunarya, “Sentiment Analysis of Online Services at the Engineering and Vocational Faculty of Ganesha Education University Using Naïve Bayes and LSTM Algorithms,” MALCOM: Indonesian Journal of Machine Learning and Computer Science, vol. 4, no. 3, pp. 1120–1129, 2024, doi: https://doi.org/10.57152/malcom.v4i3.1336.
I. G. A. P. Mahendra, I. M. A. Wirawan, and I. G. A. Gunadi, “Enhancement performance of the Naïve Bayes method using AdaBoost for classification of diabetes mellitus dataset type II,” International Journal of Advances in Applied Sciences, vol. 13, no. 3, pp. 733–742, 2024, doi: 10.11591/ijaas.v13.i3.pp733-742.
I. M. A. Wirawan, I. G. M. Darmawiguna, and I. B. N. Pascima, “Improving baseline reduction for emotion recognition based on electroencephalogram signals,” IAES International Journal of Artificial Intelligence, vol. 13, no. 4, pp. 4263–4272, 2024, doi: 10.11591/ijai.v13.i4.pp4263-4272.
F. Alrasheedi, X. Zhong, and P.-C. Huang, “Padding Module: Learning the Padding in Deep Neural Networks,” IEEE Access, vol. 11, pp. 7348-7357 2023, doi: 10.1109/ACCESS.2023.3238315.
I. A. Putra, O. D. Nurhayati, and D. Eridani, “Human Action Recognition (HAR) Classification Using MediaPipe and Long Short-Term Memory (LSTM),” Teknik, vol. 43, no. 2, pp. 190–201, 2022, doi: 10.14710/teknik.v43i2.46439.
M. El Ghazi and N. Aknin, “Optimizing Deep LSTM Model through Hyperparameter Tuning for Sensor-Based Human Activity Recognition in Smart Home,” Informatica (Slovenia), vol. 47, no. 10, pp. 109–122, 2023, doi: 10.31449/inf.v47i10.5268.
I. G. M. W. K. Widiantara, K. Y. E. Aryanto, and I. M. G. Sunarya, “Application of the Learning Vector Quantization Algorithm for Classification of Students with the Potential to Drop Out,” Brilliance: Research of Artificial Intelligence, vol. 3, no. 2, pp. 262–269, 2023, doi: 10.47709/brilliance.v3i2.3155.
K. F. Sugiantari, P. H. Suputra, L. J. E. Dewi, and G. B. P. Putra, “Pendekatan Mlp Dalam Klasifikasi Bahasa Isyarat: Analisis Jarak Euclidean Landmark Tangan,” JIKA (Jurnal Informatika), vol. 9, no. 2, p. 209, 2025, doi: 10.31000/jika.v9i2.13368.
How To Cite This :
Refbacks
- There are currently no refbacks.
