Multiple Visual Feature Integration Based Automatic Aesthetics Evaluation of Robotic Dance Motions

Hua Peng,Jinghao Hu,Huosheng Hu,Jing Li,Hui Ren,Cong Sun,Haitao Wang

doi:10.3390/info12030095

Hua Peng, Jinghao Hu + Show 5 more

Open Access

https://doi.org/10.3390/info12030095

Copy DOI

Abstract

Imitation of human behaviors is one of the effective ways to develop artificial intelligence. Human dancers, standing in front of a mirror, always achieve autonomous aesthetics evaluation on their own dance motions, which are observed from the mirror. Meanwhile, in the visual aesthetics cognition of human brains, space and shape are two important visual elements perceived from motions. Inspired by the above facts, this paper proposes a novel mechanism of automatic aesthetics evaluation of robotic dance motions based on multiple visual feature integration. In the mechanism, a video of robotic dance motion is firstly converted into several kinds of motion history images, and then a spatial feature (ripple space coding) and shape features (Zernike moment and curvature-based Fourier descriptors) are extracted from the optimized motion history images. Based on feature integration, a homogeneous ensemble classifier, which uses three different random forests, is deployed to build a machine aesthetics model, aiming to make the machine possess human aesthetic ability. The feasibility of the proposed mechanism has been verified by simulation experiments, and the experimental results show that our ensemble classifier can achieve a high correct ratio of aesthetics evaluation of 75%. The performance of our mechanism is superior to those of the existing approaches.

Highlights

As a good breakthrough point of artificial intelligence research, robotic dance is widely used to explore and develop a robot’s autonomous ability, interaction ability, imitation ability, and coordination ability with the environment [1,2,3]
To provide a solution of the core problem of autonomous robotic choreography, this paper proposed the mechanism of automatic aesthetics evaluation of robotic dance motions based on multiple visual feature integration
Using the technologies of computer vision and ensemble learning, we presented, for robotic dance motions, an automatic machine aesthetics mechanism based on multiple visual feature integration

Summary

Introduction

As a good breakthrough point of artificial intelligence research, robotic dance is widely used to explore and develop a robot’s autonomous ability, interaction ability, imitation ability, and coordination ability with the environment [1,2,3]. Any robotic dance motion has its own aesthetic attribute and constraint. If a robot perceives the aesthetics of its own dance motions just like this, it expresses more autonomous, humanoid behavior [3] and, to a certain extent, develops machine consciousness [8]. It is meaningful to explore the self-aesthetics of robotic dance motions. This paper explores the following key problem: How can a robot achieve an automatic aesthetic evaluation of its own dance motions, using only its visual information?

Methods

Results

Discussion

Conclusion