Realistic animation generation method based on Kinect

Abstract

The invention belongs to the field of computer vision and computer graphics, and provides a method which can ensure a newly-generated video sequence to be vivid and reliable and solve the problem of high requirements on an experimental device. According to the technical scheme adopted by the invention, the realistic animation generation method based on Kinect uses a single Kinect v2.0 to carry out static three dimensional model and color depth skeleton information acquisition on a person. A marker-free motion capture method is used, skeleton motion information of each frame and a three dimensional model after deformation are obtained. After a user designates a new action, on the basis of the built data collection, a video for the new action by the person is generated through the proposed sparse representation-based low rank matrix filling texture synthesis method. The method of the invention is mainly applied to video processing.

Description

technical field [0001] The invention belongs to the fields of computer vision and computer graphics, in particular to a method for generating a new action video with a strong sense of reality by using a Kinect somatosensory camera. Background technique [0002] For a long time, research on new animation generation methods has never stopped, and its solutions have emerged in an endless stream. These methods are roughly divided into three categories. The first method uses skeleton and model rendering to reconstruct the model of simple scenes and use computer rendering to obtain new animation video sequences. However, this method works well for rigid objects, but cannot form a realistic video sequence for non-rigid objects. The second method uses physical modeling to synthesize a small number of high-precision, high-realistic 3D models offline, and then uses these 3D models as samples to synthesize 3D models with realistic geometric details under various human movements. Howe...

AI Technical Summary

Problems solved by technology

However, this kind of method has very limited control over the setting of clothing structure and material parameters, and the clothing structure and material corresponding to the generated motion cannot completely match a specific clothing in reality.

The third method is based on the method of Video Textures. By extracting and reorganizing some frames in the existing video sequence set, a new video sequence is finally generated. This method can form a new animation result with a more realistic effect for simple scenes. , but cannot handle more complex scenarios

However, this scheme is expensive and requires high experimental conditions, which is not conducive to industrialization

Images