3.8 Proceedings Paper

A Skeleton Posture Transfer Method from Kinect Capture

Publisher

IEEE
DOI: 10.1109/VRHCIAI57205.2022.00033

Keywords

posture transfer; skeleton; Kinect; 3D model; virtual reality

Funding

  1. Hainan Natural Science Foundation [620RC604]
  2. Science and Technology Project of Haikou [2020-014, 2020-044, 2020-053]
  3. Guilin University of Electronic Technology, Guangxi Key Laboratory of Image and Graphic Intelligent Processing (GIIP2012)
  4. National Natural Science Foundation [61502127]
  5. Key R&D Projects in Hainan Province [ZDYF201901O]
  6. Research project on education and teaching reform of Hainan Normal University [hsjg2020-20]

Ask authors/readers for more resources

This paper proposes a method to transfer user's posture to a 3D model, by capturing skeleton information using Kinect and applying a series of transformations on the skeleton. The feasibility and effectiveness of this method have been demonstrated through experiments, and it can be applied to other types of models as well.
At present, 3D models are applied to many fields, and the action control of 3D model needs be studied further. In order to control a 3D model using user-acceptable devices, this paper proposes a method to transfer the posture of the skeleton captured by Kinect to a skeleton of the Soldier model, which is a sample in Microsoft's DirectX SDK First, using the Kinect to collect the user's skeleton information, combined with the initial posture of the Soldier skeleton, through node matching, inverse mirror transformation, transformation from absolute to relative, coordinate space transformation, to generate a skeleton with the same structure as the Soldier skeleton, which can be applied to the Soldier model. And in a virtual reality scene, the corresponding actions of users can drive the Soldier model. Experiments show that the method is feasible and effective. The proposed method can also be referenced and applied to a target skeleton with other structures.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

3.8
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available