What is it about?

Music-driven group choreography poses a considerable challenge but holds significant potential for a wide range of industrial applications. The ability to generate synchronized and visually appealing group dance motions that are aligned with music opens up opportunities in many fields such as entertainment, advertising, and virtual performances. However, most of the recent works are not able to generate high-fidelity long-term motions, or fail to enable controllable experience. In this work, we aim to address the demand for high-quality and customizable group dance generation by effectively governing the consistency and diversity of group choreographies. In particular, we utilize a diffusion-based generative approach to enable the synthesis of flexible number of dancers and long-term group dances, while ensuring coherence to the input music. Ultimately, we introduce a Group Contrastive Diffusion (GCD) strategy to enhance the connection between dancers and their group, presenting the ability to control the consistency or diversity level of the synthesized group animation via the classifier-guidance sampling technique. Through intensive experiments and evaluation, we demonstrate the effectiveness of our approach in producing visually captivating and consistent group dance motions. The experimental results show the capability of our method to achieve the desired levels of consistency and diversity, while maintaining the overall quality of the generated group choreography.

Featured Image

Why is it important?

This research holds paramount importance as it pioneers advancements in music-driven group choreography, tackling critical limitations prevalent in current methods. By introducing a diffusion-based generative approach alongside the Group Contrastive Diffusion (GCD) strategy, this work enables the creation of high-fidelity, long-term group dance motions synchronized with music, addressing the demand for customizable and visually captivating choreographies. Its significance lies in its potential impact across various industries—enhancing entertainment, advertising, and virtual performances—as well as its contribution to technological innovation by introducing novel generative techniques. Moreover, this research empowers artists and choreographers with tools for creative expression while offering audiences immersive and engaging experiences, ultimately advancing the field by demonstrating a method that effectively balances consistency, diversity, and overall quality in music-infused group choreography.


This publication represents a fascinating convergence of two worlds that often seem disparate – music and choreography. The innovative approach using diffusion-based generative techniques coupled with the Group Contrastive Diffusion (GCD) strategy truly captivates my interest. From an artistic standpoint, the potential for creating visually stunning and synchronized group dance sequences aligned with music seems incredibly promising. It opens doors for choreographers and performers to explore new realms of creativity, enabling the crafting of performances that engage audiences on a profound level. Furthermore, from a technological perspective, witnessing the advancement in generative methodologies to facilitate such synchronized and customizable dance motions highlights the ever-evolving landscape of AI-driven creativity. This intersection of technology and artistic expression not only fuels my curiosity but also sparks excitement about the endless possibilities for innovation and captivating experiences in both entertainment and creative industries.

Quang D. Tran

Read the Original

This page is a summary of: Controllable Group Choreography Using Contrastive Diffusion, ACM Transactions on Graphics, December 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3618356.
You can read the full text:




The following have contributed to this page