Temporal scaling of leg motion for music feedback system of a dancing humanoid robot

Abstract

Abstract-In this paper, we propose a method to achieve temporal scaling of leg motions as a fundamental technique for a music feedback system of a dancing humanoid robot. We asked dancers to perform dance motion at normal musical tempo and faster musical tempos and observed how dancers modified performance for given musical tempos. The obtained insights from the observation are 1) a dancer needs to preserve leg postures that are important to emphasize dance expression, 2) there is a priority to determine what features of leg motion can be adjusted, and 3) stylistic leg motion resembles normal step motion if dancers cannot follow fast musical tempo completely. Based on these insights, we generate leg motion appropriately adjusted for changing musical tempo while maintaining balance. We validated our method via simulation experiments with a humanoid robot HRP-2.

A study on leg posture recognition from Indian classical dance using Kinect sensor

“…They also used a novel head model in combination with the previous model in order to get a result of better accuracy and authenticity. In [19], the authors devise a novel technique that is based on the change in the postures and overall performance as the tempo of the music varies. A Humanoid Robot HRP-2 was used so that the method proposed could be authenticated by the usage of simulated experiments.…”

Humanized Robot Dancing: Humanoid Motion Retargeting Based in a Metrical Representation of Human Dance Styles

“…As such, key-poses most show representative instances of the motion. The techniques to analyze and determine the appropriate key-poses, in dance motion, are based on a spatial analysis of the performed body motion [5] [4] or by analyzing the dance music [10], and, in some cases, a simultaneous analysis of both aspects [16] [17] [15] [9]. Working on Latin dances [5], uses information about the main characteristics of Merengue dance style, focusing only on components of the rotation of shoulders and hips.…”

“…The performed musical analysis is similar to [16], extracting the music beat and degree of chord change for the beat structure analysis, and music intensity for mood analysis. There are also works based on temporal scaling techniques, for upper-body [15] and leg [9] motion. In [15] the dance motion is captured at different speeds and by comparison of the variance in the motion, the authors observed that some poses stay preserved.…”

“…These poses are considered key-poses since they tend to represent important moments to the music. The analysis in [9] is similar to [15], but focusing on the analysis of the step motion. The determination of the key-poses is made by using the indication that the original timings for tasks around keyposes are maintained and that stride length is also close to the original, even at different speeds.…”

“…Dance motion generation techniques typically aim at generating motion from the key-poses extracted in a prior motion analysis phase. [12], [13], [3], [9] and [7] apply Inverse Kinematics (IK) to transform the marker positions from the motion capture data into robot joint angles. In [3], the Inverse Kinematics is only applied to the upper-body, while the pelvis, leg and feet motion is generated by optimization based on the Zero Motion Point (ZMP) trajectory and dynamic mapping.…”