Tipiano: Cascaded Piano Hand Motion Synthesis via Fingertip Priors
Abstract
Synthesizing realistic piano hand motions requires both precision and naturalness. Physics-based methods achieve precision but produce stiff motions; data-driven models learn natural dynamics but struggle with positional accuracy. Piano motion exhibits a natural hierarchy: fingertip positions are nearly deterministic given piano geometry and fingering, while wrist and intermediate joints offer stylistic freedom. We present [OURS], a four-stage framework exploiting this hierarchy: (1) statistics-based fingertip positioning, (2) FiLM-conditioned trajectory refinement, (3) wrist estimation, and (4) STGCN-based pose synthesis. We contribute expert-annotated fingerings for the F\"urElise dataset (153 pieces, ~10 hours). Experiments demonstrate F1 = 0.910, substantially outperforming diffusion baselines (F1 = 0.121), with user study (N=41) confirming quality approaching motion capture. Expert evaluation by professional pianists (N=5) identified anticipatory motion as the key remaining gap, providing concrete directions for future improvement.
Keywords
Cite
@article{arxiv.2604.09692,
title = {Tipiano: Cascaded Piano Hand Motion Synthesis via Fingertip Priors},
author = {Joonhyung Bae and Kirak Kim and Hyeyoon Cho and Sein Lee and Yoon-Seok Choi and Hyeon Hur and Gyubin Lee and Akira Maezawa and Satoshi Obata and Jonghwa Park and Jaebum Park and Juhan Nam},
journal= {arXiv preprint arXiv:2604.09692},
year = {2026}
}