Researchers at the Karlsruhe Institute of Technology (KIT) and the University of Duisburg-Essen used computer-assisted neural networks to accurately identify emotions from the body language of tennis players during games. This is the first time that a model based on artificial intelligence (AI) has been trained using data from actual games. The study shows that AI can assess body language and emotions with a level of accuracy similar to that of humans. However, it also raises ethical concerns.The study focused on how AI can analyze the body language and emotions of tennis players during games. The researchers used data from real games to train a model based on artificial intelligence (AI). They found that the AI was able to accurately assess body language and emotions, similar to how humans would. However, the study also raised ethical concerns. The research was published in the journal Knowledge-Based Systems and involved collaboration between sports sciences, software development, and computer science researchers from KIT and the University of Duisburg.Essen created a specialized AI model that utilized pattern-recognition programs to analyze video footage of tennis players during live games. The success rate of their model was recorded at 68.9 percent.
Professor Darko Jekauc of KIT’s Institute of Sports and Sports Science stated, “Our model can identify affective states with an accuracy of up to 68.9 percent, which is comparable and sometimes even superior to assessments made by both human observers and earlier automated methods.”
One unique aspect of the study was the team’s use of real-life scenes, rather than simulated or contrived situations, to train their AI system.The scientists captured video clips of 15 tennis players in a specific environment, paying close attention to their body language when they either won or lost a point. The videos depicted players displaying various cues, such as bowing their heads, raising their arms in celebration, letting their racket hang, or walking at different speeds. These cues could be used to recognize the emotional states of the players. Once the AI was given this information, it learned to associate the body language signals with different emotional reactions and to determine whether the players had won a point (positive body language) or lost a point (negative body language). “Training in natural contexts is a significant advance for”Jekauc said that AI algorithms have the potential to surpass human observers in identifying real emotional states and making predictions in real scenarios. The research also found that both humans and AI are better at recognizing negative emotions than positive ones. This may be because negative emotions are expressed in more obvious ways, according to Jekauc. Psychological theories suggest that people are more sensitive to negative emotions, making them easier to identify.The ability to perceive negative emotional expressions is an evolutionary advantage, as it allows for quick resolution of conflict, which is crucial for social harmony. It is important to clarify the ethical aspects before implementing this technology. This study suggests that reliable emotion recognition could be beneficial in various areas, such as sports, healthcare, education, customer service, and automotive safety. It could improve training methods, team dynamics, performance, and prevent burnout.Jekauc emphasized the need to consider potential risks, such as privacy concerns and data misuse, despite the significant benefits of the technology. The study followed ethical guidelines and data protection regulations. It is important to address ethical and legal issues before implementing such technology in the future.
