How can AI motion models capture the nuance, expertise, and history that dance artists hold in their bodies? Improving Motion Modeling with Dance Expertise for Archiving and Analysis brings together a team of dance historians, computer scientists, and Martha Graham Technique(™) specialists to evaluate motion capture, monocular and multi-view video, and computer vision. Producing technical specifications for enhanced 2D and 3D motion modeling, the project establishes a foundation and agenda for multi-modal and AI-enabled research on dance history, choreographic preservation, and embodied legacies.

Improving Motion Modeling is funded by a 2026 Schmidt Sciences HAVI Development Award, as one of 11 interdisciplinary research teams across the globe to explore ambitious, collaborative AI interventions that have the potential to catalyze major breakthroughs in the humanities. As part of the Humanities and AI Virtual Institute (HAVI), these short-term development projects, spanning disciplines from archaeology and dance to environmental humanities and colonial history, will demonstrate how new technical approaches can be successfully implemented at scale. By bringing together computer scientists and humanists to tackle fundamental research bottlenecks, HAVI aims to both illuminate the human record and advance the design of more capable and culturally informed AI systems.

Brought together through a HAVI convening at the Sorbonne Cluster for Artificial Intelligence in September 2025, Improving Motion Modeling includes: Principal Investigators Harmony Bench (The Ohio State University, US), Ashley Brown (Martha Graham School, US), Kate Elswit (Royal Central School of Speech and Drama, UK), Michael Neff (UC Davis, US), and Michael Rau (Stanford University, US); Co-Investigator Tia-Monique Uzor (Royal Central School of Speech and Drama, UK); and Specialists Vita Berezina-Blackburn (The Ohio State University, US), and Peter Broadwell (Stanford University, US). Leveraging this interdisciplinary expertise, Improving Motion Modeling envisions performing arts knowledge and archives as transformative for technological innovation, and lays the foundation for future breakthroughs in the computational study of artistic praxis. The team kicked off this phase of research in January 2026 with a convening at the Advanced Computing Center for the Arts and Design at Ohio State.

Harmony Bench, who is Associate Professor in Ohio State's Department of Dance, described what it means to host this research, stating: “Ohio State has always been a site where dance and technology converge in exciting ways. With incredible partners at ACCAD and resources at the Motion Lab, it only made sense to bring this amazing team of researchers and artists to Ohio State. We are so lucky to have the motion capture expertise of ACCAD’s Vita Berezina-Blackburn as part of the team, and Michael Hesmond delivered an impressive last‑minute solution to a complex technical setup in the Motion Lab when our research objectives shifted unexpectedly.” Vita agreed, noting that “it is a privilege to work alongside such an exceptional, cross‑disciplinary team of leading experts, contribute to technological innovations that impact multiple fields of practice, and engage our graduate and undergraduate students in a valuable hands-on experience with advanced technology‑driven research.” The Director of ACCAD, Chris Coleman, spoke about the work in relation to the mission of the Center, reflecting how “it is critical that we leverage our research and technical expertise in collaboration with people inside and outside Ohio State to build capacity and understanding for the arts and humanities, especially around these moments when technology has blindspots about human expression.” The project team is excited to develop this research over the next several months and to see what the future of motion modeling holds.