Vocaodoru is a VR dancing game built in Unreal Engine to satisfy the Major Qualifying Project requirements at Worcester Polytechnic Institute in partnership with Ritsumeikan University in Kyoto, Japan. The game leverages integration with MikuMikuDance to create a library of songs and dances that can be imported and played in a roomscale VR environment. Vocaodoru features a dynamic limb-length calibration system that allows players with different ranges of motion to dance along to their favorite songs and be graded fairly alongside an innovative artificial intelligence that generates a music video as you play. Complete with Twitch integration, viewers can vote on their favorite shots and influence the AI's cinematographic profile over time to personalize the experience.
Vocaodoru was developed by a three-person team from Worcester Polytechnic Institute: Aidan O'Keefe, David Giangrave, and Patrick Critz. Aidan was the primary gameplay designer and programmer and co-designed the cinematography AI alongside David. Aidan developed a limb-motion based calibration system to allow players with limb differences or limited range of motion to play the game in the same way as able-bodied players. He also created an art pipeline that allowed for instant integration of MikuMikuDance "motion" files alongside a custom gameplay file format that affords custom timing and lighting information per song. Aidan helmed the team tasked with building VR integration and created the full-body inverse kinematic setup that powers the player in-game. The full thesis may be read below.
