A Harvard research team has developed BRIDGE, a simulation system that transforms standard standing-basketball footage into realistic wheelchair basketball video — giving para-athletes access to the kind of video analysis long taken for granted in non-disabled sports. The work just earned a Best Paper Award at CHI 2025.
Para-athletes studying opponents’ game tape have long faced a problem their non-disabled counterparts never consider: the footage available is almost always shot from a standing player’s perspective, requiring wheelchair basketball players to mentally rewire every movement, angle and tactic they watch. A new computer science system from Harvard University aims to eliminate that cognitive tax entirely.
Developed by Harvard’s Visual Computing Group, BRIDGE — short for “BRIDGE: Borderless Reconfiguration for Inclusive and Diverse Gameplay Experience Via Embodiment Transformation” — automatically converts broadcast basketball footage featuring non-disabled players into realistic, wheelchair-adapted video representations. The research earned a Best Paper Award at the ACM Conference on Human Factors in Computing Systems (CHI), widely considered the top venue in human-computer interaction research.
The Problem With ‘Default’ Sports Technology
Senior author Hanspeter Pfister, the An Wang Professor of Computer Science at Harvard’s John A. Paulson School of Engineering and Applied Sciences, and his team noticed something troubling while working on earlier sports-tracking tools: the systems quietly assumed a non-disabled body as the default. Co-lead author Chunggi Lee, a doctoral student in the Visual Computing Group, noted the realization pushed the team in a new direction.
“We were motivated to expand our research to inclusive sports analytics and accessible tools,” Lee said in a news release.
A key collaboration helped sharpen the research’s focus. Co-lead author Hayato Saiki, a visiting scholar from the University of Tsukuba, connected the team with Japan’s national wheelchair basketball team. What the researchers heard from those players drove home just how significant the gap was.
“Through our collaboration with the team, we realized that the main bottleneck was not tactical understanding itself, but the constant effort needed to translate stand‑up footage into wheelchair play,” Lee added. “What made it compelling was hearing national wheelchair basketball team players describe how much cognitive effort they spend mentally translating non-disabled footage.”
How BRIDGE Works
The system runs broadcast video through a reconstruction pipeline that detects and tracks players and the ball in three dimensions, generating full 3D play sequences from standard camera angles. It then applies what the researchers call an “embodiment-aware” visualization framework that separately maps head orientation, trunk movement, and wheelchair base positioning.
That layered approach is designed to communicate something nuanced: not just where a player moves, but where they are looking, what they intend to do next, and how the physical constraints of a wheelchair shape every decision on the court. The system also accounts for functional classification differences among wheelchair basketball players — a critical factor the sport uses to ensure equitable competition.
In a controlled study with 20 participants — 10 members of the Japanese national wheelchair basketball team and 10 non-elite players — BRIDGE significantly improved the naturalness of player postures in converted video and made tactical intentions clearer. Participants said the transformed footage more accurately reflected what wheelchair athletes are actually capable of, compared with unmodified non-disabled video.
Why It Matters for Inclusive Sport
The gap in sports analytics tools between disabled and non-disabled athletes is not just an inconvenience — it represents a structural inequity in access to training resources. Professional and collegiate non-disabled sports have benefited for decades from sophisticated video analysis platforms. Para-athletes, meanwhile, have largely had to make do with footage that doesn’t reflect their bodies or game.
Lee noted the project reshaped how the team thinks about design itself.
“This experience taught us to ground visualization and reconstruction methods in the real constraints of specific athlete communities, and to treat bodily differences not as edge cases, but as core design parameters for more inclusive tactical learning tools,” he added.
For college students studying computer science, human-computer interaction, or sports science, the research offers a concrete example of how technical design choices — even seemingly neutral ones — can either reinforce or break down real-world barriers.
What Comes Next
The Harvard team envisions extending the concept of “embodiment transformation” well beyond wheelchair basketball. Future directions include incorporating augmented reality, virtual reality and artificial intelligence to support other parasports, rehabilitation contexts, and even youth or older athlete populations.
Source: Harvard John A. Paulson School of Engineering and Applied Sciences