Duet dancing from a solo dance video
Year: 2025
Type: article
Abstract: This paper introduces a method to generate duet dance videos from an inputsolo dancer’s video performance. Addressing this novel problem, our systemtackles a set of sub-tasks, including dancer segmentation, camera motionhandling, stage reconstruction and the intricate management of geometricconstraints such as dancer scale preservation and dancers collision preven-tion. The proposed approach leverages both existing methodologies and newsolutions. Notably, we address collisions in 2D space-time directly, departingfrom traditional 3D approaches. We modify the initial location of the dancerto avoid long temporal collisions globally, while also modulating the pace ofthe dance by deliberate slowing down or accelerating motion to avoid shortcollisions locally. Experimental results attest to the efficacy of our approach.The system not only successfully synthesizes engaging duet dance sequencesbut also upholds the authenticity of individual performances as shown by auser study.
Source: Computers & Graphics
Authors: Yarin Moshe, Yael Moses, Ariel Shamir
Institution: Reichman University
Language: English
Topic: Human Motion and Animation
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,770
Award: G1065439166
OpenAlex ID: W4410773294