Generation of Human Walking Animation in an Arbitrary Direction Using Video Interpolation

DOI Web Site 1 Citations 5 References
  • FURUYA TADASUKE
    Graduate School of Technology, Tokyo University of Agriculture & Technology
  • SAITO TAKAFUMI
    Graduate School of Technology Bio-Applications and Systems Engineering, Tokyo University of Agriculture & Technology

Bibliographic Information

Other Title
  • 実写画像に基づく任意方向への人物歩行アニメーションの生成
  • ジッシャ ガゾウ ニ モトヅク ニンイ ホウコウ エ ノ ジンブツ ホコウ アニメーション ノ セイセイ

Search this article

Abstract

We purpose a method to create human walking animation in the arbitrary direction with little cost. We apply image-based rendering techniques for the cost reduction of human modeling. Human walking sequence along a circle is captured as input video images. A human image in each walking phase and each walking direction is extracted from the video. By interpolation of these images, the human image walking in the arbitrary direction is created. <BR>For rendering process, we use a human texture image, which is a correction of walking images during one walking sequence along several fixed direction. A human texture is stretched to billboard and placed in 3-dimensional scene. The human image displayed from the view angle is found in the human texture. Animation is created by displaying a human walking image one by one. However, by moving billboard at a fixed interval cannot show a human walk. It is not always moving to the same timing. In a human walk, the amount of movements changes by the walk posture. It is necessary to move billboard in accordance with a walk posture. <BR>We tried to create several kinds of human images from one video sequence. First, a mask image is generated by extracting the color of a shirt or trousers. Then the color of a shirt or trousers can be changed to another color. In this way we can create human walking images with various clothes color combination. Furthermore, color of hair and skin can also be changed. The human walking image was made by changing the size of the length of human image. Probably, it will be possible by replacing a human face to create much more human images. <BR>We proposed a low cost technique to generate a human walking image in arbitrary direction, with various clothes color combination. Improvement of animation quality and simulation of crowd are lets as the future studies. <BR>

Journal

Citations (1)*help

See more

References(5)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top