Framework of extrinsic feedback evaluation using motion capture for martial arts
Capturing and analysing human motion using the motion capture (MoCap) technique is a rapidly growing research area due to a large number of potential applications and its inherent complexity. In martial arts (MAs), MoCap is used widely to evaluate human performance. MoCap normally requires motion re...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Capturing and analysing human motion using the motion capture (MoCap) technique is a rapidly growing research area due to a large number of potential applications and its inherent complexity. In martial arts (MAs), MoCap is used widely to evaluate human performance. MoCap normally requires motion recognition (MoRec) to translate the m vements executed into meaningful forms. MoRec can be used to produce intrinsic and extrinsic feedback (EF) to evaluate MAs. MAs promote a sense of identity and culture, and as such are considered part of the national heritage in certain countries. However, the evaluation systems using MoCap for MAs are scarce due to the absence of a complete framework of evaluation systems for MAs. Furthermore, common approaches to MoRec require very large training and validation sets. MoRec using Gesture Description Language (GDL) and Reverse-GDL (R-GDL) approaches is insufficient to evaluate MAs techniques because both approaches are not designed to produce results in the form of EF. Therefore, a framework for evaluating EF using MoCap for MAs called EFE-MoCap has been developed. The framework comprises designs of templates, Frames-Based Rules-Range (FBRR) and Score Rubric Assessment (SRA) to create the MAs motion templates; create rulesranges for the created motion templates for publishing EF, and evaluate the framework for a traditional Malay MAs. In order to implement the framework, experimental approaches using Seni Silat Cekak Malaysia (SSCM) techniques have been carried out. The four SSCM techniques are Buah Jatuh Pertama, Buah Jatuh Ali Patah Sudah (II), Buah Jatuh Kilas Hadapan, and Buah Jatuh Hempok (II) respectively from Kaedah A, B, C, and D. By implementing the framework, the scores of the FBRR approach are consistent with the scores of SRA and the score values in each step for FBRR approach are accurate with th cor values in each step for SRA. These accuracies show the capability of the FBRR approach to detect the existence and inexistence of steps in the sequence of the techniques. Using the framework, the study has contributed to the introductions of new datasets of the motion templates for SSCM techniques, FBRR for creating the rules-ranges of the motion templates and SRA for validating and verifying the created motion templates using FBRR. |
|---|---|
| Physical Description: | xvii, 375leaves: colour illustrations; 30cm. |
| Bibliography: | Includes bibliographical references (leaves 225-243) |