Physics of Sound & Vibration

          An Inter-Disciplinary Resource Website to Effects on Human Electrodynamic Physiology

 

                                     www.uncg.edu/~t_hunter/sound.html

 Main Menu

Site Map

 

Patent No. 6621939  Scene description generating apparatus and method, object extracting method, and recording medium (Negishi, et al., Sep 16, 2003)

Abstract

A scene description generating apparatus and method, an object extracting method, and a recording medium extract an object from an input image. Positional information on the extracted object is output. Based on the positional information, scene description information about a placement position of the object in a scene is generated. When the object is deformed, reference to the positional information is made, and the scene description information is generated in which the object deformation is reflected. Accordingly, the object is placed at a desirable position in the scene.

Notes:

SUMMARY OF THE INVENTION

Accordingly, it is an object of the present invention to provide a scene description generating apparatus and method and an object extracting method for solving the above problems, that is, for preventing generation of undesirable shifting or distortion in a scene described by a scene description even when an object in an input image or graphic data is deformed, and for reflecting movement of the object in the input image or the graphic data in movement of the object or in movement of the texture in the scene.

According to an aspect of the present invention, the foregoing objects are achieved through provision of a scene description generating apparatus and method including an object extracting step of extracting an object from an input image and outputting positional information on the extracted object. Based on the positional information output in the object extracting step, scene description information about a placement position of the object in a scene is generated in a scene description generating step. When the object is deformed, the positional information is referred to in the scene description generating step and the scene description information in which the object deformation is reflected is generated.

According to another aspect of the present invention, the foregoing objects are achieved through provision of a scene description generating apparatus and method including an object extracting step of extracting an object from an input image. In a positional information detecting step, positional information on the object extracted in the object extracting step is detected. Based on the positional information detected in the positional information detecting step, scene description information about a placement position of the object in a scene is generated in a scene description generating step. When the object is deformed, the positional information is referred in the scene description generating step and the scene description information in which the object deformation is reflected is generated.

According to another aspect of the present invention, the foregoing objects are achieved through provision of a recording medium for causing a scene description generating apparatus for generating scene description information on an object to execute a computer-readable program. The program includes an object extracting step of extracting the object from an input image and outputting positional information on the extracted object. Based on the positional information output in the object extracting step, the scene description information about a placement position of the object in a scene is generated in a scene description generating step. When the object is deformed, the positional information is referred to in the scene description generating step and the scene description information in which the object deformation is reflected is generated.

According to the present invention, when placing an object segmented from a static image signal, a moving image signal, or graphic data by an object extracting unit/step in a screen and describing a new scene, the object extracting unit, i.e., a segmentation unit, outputs positional information on a region containing the object in the input image or the graphic data. Based on the output positional information, a scene description generating unit/step determines a placement position of the object. Accordingly, even when the region containing the object is deformed or shifted, the object is placed at a desirable position in the scene described by the scene description. When the segmented object is used as a texture in the scene description, the scene description is generated in which texture coordinates are transformed based on the positional information output from the segmentation unit. Therefore, distortion of a texture pasted in the scene is prevented, and shifting of the object is reflected in the texture. Alternatively, texture distortion is prevented by changing the size of a scene object on which the texture is to be pasted or by changing the position of the texture.

When the positional information on the region containing the object in the image or the graphic data is included in data of the segmented object, the positional information is made equally available by means of a positional information detector to which the object data is input to detect the positional information. Hence, undesirable shifting or distortion in the scene is prevented.

When the region is determined so as to contain objects in frames of a plurality of images or graphic data and is segmented, the number of changes of the placement position is reduced, or changes are not necessary at all. In particular, when the region containing the object is set as a picture frame of the input image or the graphic data, it is not necessary to change the placement position.