CustomDancer: Customized Dance Recommendation by Text-Dance Retrieval

TL;DR

CustomDancer is a multimodal framework that improves personalized dance retrieval by aligning text, music, and motion data, supported by a new large-scale dataset.

Contribution

It introduces TD-Data, a comprehensive dance dataset, and proposes CustomDancer, a novel retrieval model that enhances text-based dance search accuracy.

Findings

CustomDancer achieves 10.23% Recall@1 on TD-Data.

The framework improves retrieval quality in benchmarks.

User studies favor CustomDancer over baselines.

Abstract

Dance serves as both a cultural cornerstone and a medium for personal expression, yet the rapid growth of online dance content has made personalized discovery increasingly difficult. Text-based dance retrieval offers a natural interface for users to search with choreographic intent, but it remains underexplored because dance requires simultaneous reasoning over linguistic semantics, musical rhythm, and full-body motion dynamics. We introduce TD-Data, a large-scale open dataset for text-dance retrieval, containing about 4,000 12-second dance clips, 14.6 hours of motion, 22 genres, and annotations from professional dance experts. On top of this dataset, we propose CustomDancer, a multimodal retrieval framework that aligns text with dance through a CLIP-based text encoder, music and motion encoders, and a music-motion blending module. CustomDancer achieves state-of-the-art performance on TD-Data, reaching 10.23% Recall@1 and improving retrieval quality in both quantitative benchmarks and user preference studies.