Author(s): YASUYUKI, N., Institution: NIIGATA UNIVERSITY OF HEALTH AND WELFARE, Country: JAPAN, Abstract-ID: 576

INTRODUCTION:
The study aims to explore the disparities and characteristics in how referees perceive a basketball game. This has been achieved by analyzing utterances of referees who watched videos captured with fixed video cameras and those recorded with 360-degree omnidirectional cameras.
METHODS:
The study involved two certified referees from the Japan Basketball Association. A college basketball practice game was recorded using a fixed video camera (equipped with a wide-angle lens covering all players) and a 360-degree omnidirectional camera. These cameras were positioned at three locations: on the bench side (Point A), on the side opposite the bench (Point B), and under the goal (Point C). The two referees observed videos captured by the fixed video cameras (referred to as “fixed videos”) and the videos recorded with the 360-degree omnidirectional cameras (“VR videos”) while making utterances. Referee A initially watched the fixed videos in the sequence of Point A, Point B, and Point C, while Referee B began with the VR videos, following the order of Point C, Point B, and Point A. The referees had the freedom to express their thoughts while reviewing the recorded game videos.

RESULTS:
The utterances were categorized into six groups: “situation description,” “prediction,” “retrospect,” “foul judgment,” “question,” and “delegation (delegating to another referee’s judgment).” The total number of utterances was greater when reviewing fixed videos compared to VR videos. By category, the number of utterances in the “situation description” category was higher when watching fixed videos than when viewing VR videos. This is likely due to fixed videos being observed more objectively from a bystander’s perspective compared to VR videos. The number of utterances in the “question” category was also higher while watching fixed videos compared to VR videos. This is probably because viewers cannot control the focus at their own discretion during fixed video viewing, unlike VR video viewing, resulting in more obscured elements. In contrast, the number of utterances in the “foul judgment” category was higher when viewing VR videos than when watching fixed videos. This is likely because the VR videos more accurately represented the actual game. By location, the number of utterances in the “situation description” and “prediction” categories was higher at Points A and B than at Point C. This is likely because viewers can observe all the players from Points A and B, enabling them to perceive the game more objectively as bystanders. In the “foul judgment” category, more utterances were made at Point C than at Points A and B during the viewing of VR videos. Given that many violations occur under the goal (Point C) in actual games, it is probable that this area can be observed in VR videos in a situation similar to that in real games.
CONCLUSION:
The following two conclusions were drawn from the results and discussion:
- While watching fixed videos, viewers tend to perceive events objectively, adopting the perspective of bystanders rather than that of referees.
- While watching VR videos, viewers tend to make decisions as referees in situations resembling those in actual games.