Peripheral Raises $8.7M for Sports Spatial Intelligence

Peripheral's system takes video feeds captured from different angles and turns them into a three-dimensional environment that users can navigate. According to The Logic, Peripheral pitches the system to leagues and stadiums as a way for users to move through a 3D environment, follow particular players, or view an action from different perspectives, with fewer cameras than some alternative approaches.

This class of system combines multi-view geometry, camera calibration, neural rendering, and temporal reconstruction. In comparable deployments, the difficult engineering work is not only achieving a visually convincing reconstruction, but doing so with sufficiently low latency and stable player geometry for replay, analytics, and coaching use cases. Broadcast and venue operators also need camera placement, synchronization, compute, and rights workflows that can operate reliably in live production.

BetaKit reports that Peripheral's neural-rendering approach draws on robotics and autonomous-vehicle perception techniques. Cui and Khan previously worked in areas including self-driving systems, Huawei, Tesla, and University of Toronto Formula Racing, according to Sports Business Journal and the company's funding announcement.

Peripheral demonstrated its technology to NBA and WNBA officials at the 2026 NBA Summer League through the NBA Launchpad program, according to the company's announcement. The company said videos of the technology published through Toronto Raptors social accounts received millions of views.

Sports Business Journal reports that the financing is intended to support engineering hiring and expansion through potential league, stadium, and media customers. The Logic additionally reports that Peripheral is pitching its system as an engagement tool and is set to launch a consumer replay service this year.

Deloitte Ventures managing director Jon Wolkin said in the funding announcement that sports media is seeing momentum toward more immersive experiences, and that Peripheral's volumetric-capture approach has the potential to make high-quality immersive content more practical for leagues, broadcasters, and rights holders. Inovia partner Karamdeep Nijjar described the company's product as delivering real-time volumetric video and biomechanics at scale.

For ML and computer-vision practitioners, the funding highlights continued commercial interest in reconstruction models that turn sparse or heterogeneous visual inputs into interactive spatial representations. Comparable sports deployments will be judged less by a single replay demo than by reconstruction quality under occlusion, motion blur, changing lighting, and the latency and operating cost required for live use.