AI and Psychology Join Forces to Unlock Boxers' Behavior
-
Research backgrounds and motivations Boxing is a highly technical and psychological sport; sports psychology plays an important role in analyzing athletes' behavior and performance. AI technologies like CNN have been introduced into boxing behavior recognition.
-
Research objectives Establish a boxing behavior recognition model based on AI-CNN, explore the influence of athletes' psychological state on the accuracy of behavior recognition. Innovations include studying psychological factors, collecting high-quality behavior data, and fusing psychological text and video image data.
-
Literature review Studies show the importance of psychological factors like anxiety and concentration in boxing. Machine learning and CNN have advantages in sports behavior recognition and analysis. However, research combining sports psychology and AI is lacking.
-
Research model Survey boxers with a psychological scale covering factors like anxiety and confidence. Construct model combining BERT text processing and 3D-RESNET spatio-temporal video feature extraction for behavior recognition. Evaluate performance compared to other algorithms.
-
Conclusion and future works Proposed model shows superior performance. Integrating sports psychology provides deeper understanding of behavior recognition. Future works can expand samples and algorithms and extend the methodology to other sports.