Sports Video Athlete Detection Based on Associative Memory Neural Network

Area of Science:

• Computer Vision
• Sports Analytics
• Machine Learning

Background:

• Automatic athlete detection in sports videos is crucial for performance analysis.
• Existing methods may struggle with accuracy and real-time processing.

Purpose of the Study:

• To propose an automatic detection method for athletes in sports videos.
• To enhance the accuracy and real-time performance of athlete detection.

Main Methods:

• Utilized an adaptive motion neural network (AMNN) for athlete detection.
• Extracted moving areas, used color and texture information to identify the stadium and eliminate shadows/noise.
• Trained a neural network (NN) classifier with athlete and non-athlete images.
• Employed image pyramids and local search for precise athlete localization.

Main Results:

• The system accurately detects the motion shape of moving targets.
• Achieved real-time processing capabilities.
• Demonstrated good overall real-time performance in experiments.

Conclusions:

• The proposed AMNN-based method offers an effective solution for automatic athlete detection in sports videos.
• The system exhibits robust performance in terms of accuracy and speed.
• This approach has potential applications in sports video analysis and real-time monitoring.