Unlock instant, AI-driven research and patent intelligence for your innovation.

A real-time hair segmentation method and system

A hair and segmentation mask technology, applied in image analysis, image enhancement, instruments, etc., can solve the problems of slow processing speed, time-consuming and laborious, and achieve the effect of improving speed and ensuring real-time performance.

Active Publication Date: 2021-08-13
北京中科深智科技有限公司
View PDF23 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, at present, the above-mentioned image segmentation process is usually carried out manually. Artists need to perform artificial image segmentation and image processing on each video frame according to the requirements. Faced with a video with a large number of frames, the process of manual image segmentation is undoubtedly time-consuming. quite strenuous
[0003] So in order to solve this problem, for example, when shooting a movie, artificially arrange an easy-to-handle background, such as shooting a movie scene with a green curtain as the background, and then use special video processing software to replace the green background with the required background. The processing method also has considerable limitations. If the foreground structure is complex, such as when there is a large area of ​​hair in the foreground, the processing speed of this foreground and background segmentation method is very slow, and it cannot achieve real-time segmentation of hair

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
  • A real-time hair segmentation method and system
  • A real-time hair segmentation method and system
  • A real-time hair segmentation method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and through specific implementation methods.

[0050] Wherein, the accompanying drawings are only for illustrative purposes, showing only schematic diagrams, rather than physical drawings, and should not be construed as limitations on this patent; in order to better illustrate the embodiments of the present invention, some parts of the accompanying drawings will be omitted, Enlargement or reduction does not represent the size of the actual product; for those skilled in the art, it is understandable that certain known structures and their descriptions in the drawings may be omitted.

[0051]In the drawings of the embodiments of the present invention, the same or similar symbols correspond to the same or similar components; , "inner", "outer" and other indicated orientations or positional relationships are based on the orientations or positional r...

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 present invention discloses a real-time hair segmentation method and system, wherein the method includes: acquiring the current video frame image; separating the current video frame image into three channel images in RGB color mode; dividing the previous video frame of the current frame The first hair segmentation mask associated with the image and the three channel images associated with the current video frame image are input into a pre-trained hair segmentation model, and the current video frame image is output through model prediction. The associated second hair segmentation mask; repeat the above steps to complete the real-time hair segmentation of all video frame images. In the present invention, the three-channel image of one frame of original image and the hair segmentation mask model corresponding to the original image are input, and the hair segmentation mask of the current frame is predicted and output, which greatly improves the speed of hair segmentation and ensures the real-time performance of hair segmentation.

Description

technical field [0001] The invention relates to the technical field of image segmentation, in particular to a real-time hair segmentation method and system. Background technique [0002] Image segmentation is a widely used technique that enables film directors and video content creators to separate the foreground of an image from its background and treat them as two distinct visual layers. By modifying the layer (such as recoloring, masking or replacing, etc.), the image scene that is originally boring or does not meet the visual requirements becomes more eye-catching and improves the visual effect. However, at present, the above-mentioned image segmentation process is usually carried out manually. Artists need to perform artificial image segmentation and image processing on each video frame according to the requirements. Faced with a video with a large number of frames, the process of manual image segmentation is undoubtedly time-consuming. Quite laborious. [0003] So in...

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): G06T7/11G06T7/194G06T7/90G06N3/04
CPCG06N3/04G06T2207/10016G06T2207/10024G06T2207/20081G06T2207/20084G06T7/11G06T7/194G06T7/90
Inventor 不公告发明人
Owner 北京中科深智科技有限公司