EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Efficiency of Temporal Convolutional Networks in Karate Kata Evaluation: A Comparative Study

Viona Zatil Aqmar Kaleb, Wiranto Herry Utomo

Informatika Mulawarman : Jurnal Ilmiah Ilmu Komputer · 29.06.2026 · 21 (1) · 13

DOI: 10.30872/jim.v21i1.28888 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Informatika Mulawarman : Jurnal Ilmiah Ilmu Komputer, online erschienen am 29.06.2026. Verfasst von Viona Zatil Aqmar Kaleb und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The objective quantification of complex martial-arts movement is a long-standing challenge in computer vision. This study reports a feasibility prototype that pairs MoveNet Lightning pose estimation with a lightweight Temporal Convolutional Network for two tasks on a small custom dataset of four foundational Karate Kata (Heian Shodan through Heian Yondan): (1) multi-class kata recognition and (2) binary correctness evaluation. A 132-dimensional kinematic feature vector is constructed per frame from 17 MoveNet keypoints, combining hip-centered normalized coordinates, confidence scores, joint angles, bone-length ratios, and end-effector velocities. The full dataset comprises 100 lateral-view videos (80 train/10 validation/10 test) collected from a single Indonesian dojo. A multi-task model with two causal-dilated 1D-convolution layers (32 filters, dilations 1 and 2) is trained for up to 100 epochs on Apple M1 hardware over five random seeds. The kata head reaches 100% validation accuracy with zero variance across all five seeds; however, this result is interpreted with strong reservations because the validation set contains only 10 samples. The correctness head consistently collapses to majority-class prediction (60.0% ± 0.0%, identical to the all-positive baseline of 6/10). The contribution of this paper is therefore a fully reproducible end-to-end pipeline (MoveNet → 132-D features → multi-task TCN → real-time webcam demo) together with a candid characterization of the dataset-driven limits that block the correctness task at this scale.

Zitieren

APA

Kaleb, V. Z. A., & Utomo, W. H. (2026). Efficiency of Temporal Convolutional Networks in Karate Kata Evaluation: A Comparative Study. Informatika Mulawarman : Jurnal Ilmiah Ilmu Komputer, 21(1), 13. https://doi.org/10.30872/jim.v21i1.28888

Vancouver

Kaleb VZA, Utomo WH. Efficiency of Temporal Convolutional Networks in Karate Kata Evaluation: A Comparative Study. Informatika Mulawarman : Jurnal Ilmiah Ilmu Komputer. 2026;21(1):13. doi:10.30872/jim.v21i1.28888

BibTeX
@article{Kaleb2026,
  author = {Kaleb, Viona Zatil Aqmar and Utomo, Wiranto Herry},
  title = {Efficiency of Temporal Convolutional Networks in Karate Kata Evaluation: A Comparative Study},
  journal = {Informatika Mulawarman : Jurnal Ilmiah Ilmu Komputer},
  year = {2026},
  volume = {21},
  number = {1},
  pages = {13},
  doi = {10.30872/jim.v21i1.28888}
}
Kostenlos starten →