Related Experiment Video
Updated: Jun 21, 2026
07:12
A Gaze-Contingent Display Framework for Perceptual Learning Research with Simulated Central Vision Loss
Published on: April 11, 2025
Cross-View Multimodal Vision-Based Assessment Framework for Traditional Chinese Medicine Rehabilitation Training.
Summary
This study introduces CME-AQA, a novel vision-based system for assessing Traditional Chinese Medicine (TCM) rehabilitation training. The framework improves accuracy in acupuncture and Tuina techniques, making training more effective.
Area of Science:
- Computer Vision
- Rehabilitation Medicine
- Traditional Chinese Medicine (TCM)
Background:
- Vision-based assessment offers cost-effective evaluation for Traditional Chinese Medicine (TCM) rehabilitation.
- Existing automatic action quality assessment (AQA) frameworks struggle with TCM techniques due to self-occlusion and complex interactions.
Purpose of the Study:
- To develop an advanced vision-based assessment framework for TCM rehabilitation training.
- To enhance the accuracy and robustness of automatic action quality assessment (AQA) in TCM practices like acupuncture and Tuina.
Main Methods:
- Proposed CME-AQA: a cross-view, multimodal framework integrating visual-pose fusion.
- Utilized synchronized first-person and third-person videos for robust training.
- Collected dual-view datasets (TCMAQA61-A for Acupuncture, TCM-AQA61-T for Tuina) with expert annotations.
Main Results:
- Achieved superior or comparable performance against baselines.
- Demonstrated over 10% relative improvement in weighted F1 for key rating tasks (e.g., Needle Depth).
- Reduced mean absolute error in quantitative measures (e.g., insertion time).
Conclusions:
- CME-AQA enhances assessment accuracy for structured TCM rehabilitation training.
- The framework facilitates more convenient and effective training-oriented skill evaluation.
- Applicable to related simulated clinical skill assessments involving participant motion.