Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
2000-12-02
2004-03-16
Kelley, Chris (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C382S236000
Reexamination Certificate
active
06707851
ABSTRACT:
DESCRIPTION
1. Technical Field
The present invention relates to a video object segmentation method for correctly segmenting a video into a plurality of video objects by a user's assistance and an object tracking, when a video object-oriented manipulation, an editing, an object-based user interaction, an object-based video coding and the like are performed in an apparatus such as a video editing, creating and compositing apparatus, and an object-based video codec (or encoder/decoder).
2. Background Art
According to conventional video object segmentation methods, objects in image are automatically segmented based on the determination whether the brightness values of consecutive images are different or not, or are manipulatively segmented for every frame by user using an editing tool.
In the conventional video object segmentation methods, the automatic segmentation method causes the problem in that the segmentation capability is varied in accordance with the threshold value on which it is determined that the brightness value between the consecutive images is changed, and the manipulative segmentation method has the problem in that it takes a considerably long time because the segmentation of every frame requires the user's direct manipulation work, thus it endows the user with the economical burden.
Accordingly, the present invention is devised for solve the above problems. The objective of the present invention is to provide a video object segmentation method for correctly segmenting a video object, in which the video objects to be firstly appeared in an image sequence are segmented in manual or in semi-manual, and then the following video frames are automatically segmented by an object-tracking based on the moving of the primarily segmented video object, for the sake of detecting an object from an image in a video system.
DISCLOSURE OF INVENTION
To accomplish the objective of the present invention, there is provided a video object segmentation method applicable to a video system, comprising the steps of: primarily segmenting an object existing in a frame of video sequence manually or semi-manually; and secondly segmenting the object within a video sequence including the primarily segmented object automatically.
In one preferred embodiment of the present invention, the video object segmentation method further includes the steps of: determining whether any scene change is made between consecutive frames or any new object other than the primarily segmented object appears within the frame being automatically segmented, when repeatedly performing the step of secondly segmenting for consecutive frames; and repeating the step of primarily segmenting, if the answer of the step of determining is positive. This method also includes the steps of examining the quality of automatically segmented results,; repeating the step of secondly segmenting, if the quality of automatically segmented results is satisfactory; and repeating the step of primarily segmenting, if the quality of automatically segmented results is not satisfactory.
In addition, the step of primarily segmenting can be made by segmenting the object within the image in completely manual using an user interface tool, or by segmenting the object within the image in semi-manual, such that, if the user designates manually a rough boundary line of the object within the image, then the object within the image is automatically segmented based on the designation-related information and an image segmentation information. The step of secondly segmenting can be made by tracking the object region to which in the current frame the primarily segmented video object in the previous frame is moved, so as to segment the object within the frame of the consecutive frames.
REFERENCES:
patent: 5852673 (1998-12-01), Young
patent: 5969755 (1999-10-01), Courtney
patent: 6137913 (2000-10-01), Kwak et al.
Ahn Chieteuk
Choi Jae Gark
Kim Mun-churl
Kwak Jin Suk
Lee Myoung Ho
Blakely & Sokoloff, Taylor & Zafman
Electronics and Telecommunications Research Institute
Kelley Chris
Vo Tung Tu
LandOfFree
Method for objects segmentation in video sequences by object... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for objects segmentation in video sequences by object..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for objects segmentation in video sequences by object... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3255462