Editing method for human motions in videos

A technology in human motion and video, applied in image data processing, instruments, genetic models, etc., can solve the problems of unable to expand human body editing, unable to edit human body posture, unable to meet the requirements of appearance and posture, etc.

Inactive Publication Date: 2014-07-09
NANJING UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For example, document 1: SCHOLZ, V., EL-ABED, S., SEIDEL, H.-P., ANDMAGNOR, M.A. Editing object behavior in video sequences. CGF, 2009.28 (6): 1632–1643 realized some basic video editing methods, such as moving Deletion and addition of objects, motion editing of objects, non-rigid object deformation, key frame interpolation, and processing of camera motion, etc., and then such operations can only achieve basic editing of videos, and cannot meet the requirements such as editing the appearance and posture of the human body; literature 2: WANG, J., DRUCKER, S.M., AGRAWALA, M., ANDCOHEN, M.F. The cartoon animation filter. ACMTOG, 2006, 25(3): 1169–1173. A method of animation filtering is proposed, which can make the human body in the animation The movement becomes more exaggerated and dynamic, but this method only changes the appearance of the human body in the image through the deformation method, and it is difficult to achieve precise control of the posture; Literature 3: LEYVAND, T., COHEN-OR, D., DROR, G. ,ANDLISCHINSKI,D.Data-drivenenhancementoffacialattractiveness.ACMTOG,2008,27(3):1–9 Learn a two-dimensional deformation function from the face image training set, which can deform the face in the input image to increase the attractiveness of expressions power, but this method can only deal with face images, and cannot be extended to such as human body editing
In addition, research on human appearance editing in images and videos has also appeared in recent years, such as literature 4: ZHOU, S., FU, H., LIU, L., COHEN-OR, D., ANDHAN, X. Parametric shaping of human bodies in images. ACMTOG, 2010, 29(4): 1-10 introduces a parametric 3D human body model, and realizes direct editing of the appearance of the human body by adjusting the appearance attributes of the human body; Document 5: ARJUN Jain, Thorsten Thorm¨ahlen, Hans-Peter Seideland Christian Theobalt.Movie Reshape: Tracking and Reshaping of Humans in Videos .ACMTOG, 2010, 29(5) extended the work of document 4 to video, and realized the editing of the appearance of the human body in the video, but this method cannot edit the posture of the human body
[0004]To sum up, the current research on video editing methods mainly focuses on basic motion editing, motion special effects processing, specific object editing, etc., and the editing of the human body is limited to the editing of the appearance of the human body , research on the editing of human pose and motion in video is not yet visible

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Editing method for human motions in videos
  • Editing method for human motions in videos
  • Editing method for human motions in videos

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] The invention discloses a scheme for gesture modeling and editing and motion generation by using hand-drawn sketches, including the following steps:

[0049] Step 1, human body posture analysis: adopt the hierarchical posture optimization method based on immune evolution, calculate the human body posture of each frame image in the input video, and obtain the posture data of human body movement;

[0050] Step 2, attitude data editing: edit the posture data of human body movement obtained in step 1 by sketching interactively, and generate new posture data of human movement;

[0051] Step 3, image deformation: take the new human body motion posture data in step 2 as the control condition, and use the model-driven moving least squares image deformation method to deform each frame image in the input video to generate a new human body motion image. video.

[0052] More specifically, the present invention discloses a method for editing human body motion in video, which suppor...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a human body motion editing method in a video, which supports users to edit the human body motion in the video to obtain a video containing new human body motion, specifically including three steps of human body posture analysis, posture data editing, and image deformation: first Using the hierarchical posture analysis method based on immune evolution, the posture data of the human body movement in the input video is calculated through global posture optimization, posture sequence smoothing, and local posture optimization; secondly, the posture data of the human movement obtained by the above calculation is edited by sketching interactive mode Data to generate new posture data of human body motion; finally, each frame image of the input video is deformed by using the model-driven moving least squares image deformation method to generate a video containing new human motion. The invention can realize the editing of the human body motion in the video to obtain a video containing new human body motion, and has important applications in the fields of film post-production, advertisement design and the like.

Description

technical field [0001] The invention relates to a human body motion editing method in a video, which belongs to the research field of computer image and video processing, in particular to a method for supporting users to edit human body motion in a video. Background technique [0002] Image editing technology has been widely used in real life. For example, editing tools such as Photoshop can be used to perform various editing operations on images such as tone adjustment, structure adjustment, image repair, image denoising, and image deformation. However, existing video editing tools such as Adobe's Effects and Apple's Shake can only achieve basic operations such as image cutout and foreground extraction. It is still a very challenging task to realize the editing of video content, such as the editing of the appearance and motion of characters in the video. However, video editing has important applications in advertising design, film post-production and other fields. The pote...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/20G06N3/12
Inventor 孙正兴李毅刘凯周俊昊
Owner NANJING UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products