Jerry Lu MFin '24 and Professor Anette "Peko" Hosoi from the MIT Sports Lab have developed an innovative optical tracking system called OOFSkate that leverages artificial intelligence to enhance figure skating performance. This system aims to help skaters perfect their jumps, including the highly challenging quadruple axels and potentially even quintuple jumps. According to MIT News AI, the technology analyzes video footage of skaters' jumps and provides detailed feedback to improve rotation speed and jump height.
The development of OOFSkate addresses a critical need in figure skating, where athletes must execute complex maneuvers with precision. Elite skaters are constantly seeking ways to optimize their performance, often by increasing the number of rotations in their jumps. Landing a quintuple jump, or quint, represents the pinnacle of achievement in the sport, pushing the boundaries of human athletic capability. The MIT team's focus on using AI to support skaters aligns with the ongoing efforts to refine techniques and achieve new heights in competitive figure skating.
OOFSkate operates by analyzing video recordings of a skater's jump. The system breaks down the motion into precise metrics such as jump height and rotation speed. These measurements are compared against the performances of current and former elite athletes, providing a benchmark for improvement. According to MIT News AI, users simply record a jump, upload the video, and receive immediate feedback on their technique. The system also offers an automated grading feature, simulating how an international panel of judges might score the performance based on current standards.
The core technology behind OOFSkate relies on advanced pose estimation algorithms, which accurately track the skeletal configuration of a skater in mid-air. By focusing on the two-dimensional plane rather than depth, the system effectively captures essential elements like jump height and rotation count. This approach minimizes errors typically associated with depth perception issues in pose estimation, ensuring reliable and actionable insights for skaters.
The implications of OOFSkate extend beyond individual athlete performance improvements. As skaters refine their techniques using this technology, the potential exists for significant advancements in the sport itself. Achieving a quintuple jump would represent a groundbreaking milestone, demonstrating the limits of human athletic ability and innovation. Additionally, the system's ability to provide detailed analysis and grading can aid coaches and trainers in developing more effective training programs tailored to each athlete's unique strengths and weaknesses.
Looking ahead, the MIT Sports Lab plans to expand its research into evaluating the aesthetic components of figure skating through AI. According to MIT News AI, this new project, funded by a MIT Human Insight Collaborative (MITHIC) grant, will explore how AI can assess artistic performance. This initiative could lead to more objective and consistent evaluations of the subjective aspects of figure skating, enhancing the overall fairness and transparency of the sport.
As the 2026 Winter Olympics approach, the role of technology in supporting and advancing figure skating will likely grow more prominent. Jerry Lu's involvement with NBC Sports underscores the broader impact of these innovations, offering enhanced insights and explanations to viewers and commentators alike. The future of figure skating may indeed include AI-driven advancements that propel athletes towards achieving previously unimaginable feats.
---
Source: [MIT News AI](https://news.mit.edu/2026/3-questions-using-ai-help-olympic-skaters-land-quint-0210)