# Separate to Collaborate: Dual-Stream Diffusion Model for Coordinated Piano Hand Motion Synthesis

Source: [Separate to Collaborate: Dual-Stream Diffusion Model for Coordinated Piano Hand Motion Synthesis](<https://www.sota2.com/research/paper/68fa7b241e209131cea86d12>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T05:33:44.062Z

Authors: Zihao Liu, Mingwen Ou, Zunnan Xu, Jiaqi Huang, Haonan Han, Ronghui Li, Xiu Li

Published: 2025-04-14T05:17:41.000Z

## Abstract

Automating the synthesis of coordinated bimanual piano performances poses significant challenges, particularly in capturing the intricate choreography between the hands while preserving their distinct kinematic signatures. In this paper, we propose a dual-stream neural framework designed to generate synchronized hand gestures for piano playing from audio input, addressing the critical challenge of modeling both hand independence and coordination. Our framework introduces two key innovations: (i) a decoupled diffusion-based generation framework that independently models each hand's motion via dual-noise initialization, sampling distinct latent noise for each while leveraging a shared positional condition, and (ii) a Hand-Coordinated Asymmetric Attention (HCAA) mechanism suppresses symmetric (common-mode) noise to highlight asymmetric hand-specific features, while adaptively enhancing inter-hand coordination during denoising. Comprehensive evaluations demonstrate that our framework outperforms existing state-of-the-art methods across multiple metrics. Our project is available at https://monkek123king.github.io/S2C\_page/.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Piano Motion Generation on PianoMotion10M short-sequence 1.0](<https://www.sota2.com/research/sota/piano-motion-generation-on-pianomotion10m-short-sequence-1-0.md>): Piano Motion Generation; PianoMotion10M short-sequence 1.0; FGD (Left Hand): 0.309.

- [Piano Motion Generation on PianoMotion10M long-sequence setting](<https://www.sota2.com/research/sota/piano-motion-generation-on-pianomotion10m-long-sequence-setting.md>): Piano Motion Generation; PianoMotion10M long-sequence setting; Left Hand FGD: 0.443.

- [Piano Playing Motion Generation on FürElise Dataset (test)](<https://www.sota2.com/research/sota/piano-playing-motion-generation-on-f-relise-dataset-test.md>): Piano Playing Motion Generation; FürElise Dataset (test); Position Error (mm) Fingertip Left: 74.1.

- [Piano Key Contact Prediction on FürElise Dataset (test)](<https://www.sota2.com/research/sota/piano-key-contact-prediction-on-f-relise-dataset-test.md>): Piano Key Contact Prediction; FürElise Dataset (test); Precision: 10.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2504.09885>)

[PDF](<https://arxiv.org/pdf/2504.09885.pdf>)