JPH11213139A - Fitting simulation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To form an image more naturally without giving an unnatural feeling with a simple structure by painting out the background of a second image with a decided and fixed color and synthesizing a face of a person and the background around it in a first image with the second image painted out with the fixed color. SOLUTION: The person is photographed by using a screen 36 as the background to store in a capture I/F 28 as the first image 50. Color information is selected from plural preset sampling points of the first image 50, an average value of the color information is calculated and defined as the fixed color 74. An image of the person to select and put on clothes is defined as the second image 54. The second image 54, first and second mask data 58, 66 corresponding to the second image 54 are read from a hard disk 20 and stored in a RAM 18. The background and the face of the second image 54 are painted out with the fixed color 74 decided by a fixed color deciding means 42 by using the first mask data 58 by an image synthesizing means 44. A face part, etc., is clipped from the first image 50 and synthesized with a corrected second image 62 by the image synthesizing means 44.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a clothing try-on simulation system.

[0002]

2. Description of the Related Art An image of a person performing a fitting simulation is taken by a video camera, and a model of a model wearing a desired design is selected from among images of a model wearing various design clothes registered in advance. An image of a model wearing clothes by selecting an image and displaying it on a monitor, cutting out a face part from an image of a person taken with a video camera, and combining it with an image of a model wearing clothes of a desired design A fitting simulation system of a type that replaces an image of a face of a person who has taken a face portion of a person is known.

Conventionally, in such a try-on simulation system, chroma key processing is used as an image processing method for cutting out only a face image from an image of a person photographed by a video camera by contour extraction. In this chroma key processing, only a specific hue portion is extracted from the hue component of an image to generate a key signal, and a portion having a hue different from the key signal can be cut out. Normal,
When image processing is performed by chroma key processing, a person is photographed with a blue-based color having the least effect on skin color as a background.

[0004]

However, in the conventional try-on simulation system, there are some problems in using a chroma key to cut out the face from the image of the person performing the try-on simulation. Usually, the fitting simulation system is often installed in a store such as a boutique. Unlike the shooting studio, the interior of the store does not have a shooting environment such as lighting. It is also necessary to prepare a blue screen for the background in order to perform the chroma key processing. Further, in the chroma key processing, an advanced image processing technique is required to naturally cut out the image of the face including the hair and the like, and the cost for the apparatus is correspondingly increased. When the hair is dyed in a blue color, reflection occurs in that part.

[0005] An object of the present invention is to cut out a face portion of an image of a person photographed by a video camera or the like and replace it with a face portion of an image of a model wearing desired clothes. An object of the present invention is to provide a try-on simulation system capable of synthesizing images more naturally and without a sense of incongruity with a simple configuration without performing extraction.

[0006]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention provides an image input means for photographing a person against a plain screen as a background, inputting the image as a first image, and storing the first image. A fitting simulation system that combines the face of the model of the second image with the face of the person of the first image photographed by the image input means by combining with the second image that is the image of the model wearing the clothes stored in A fixed color determination unit that collects color information of the background of one image to determine a fixed color, and fills the background in the second image with the fixed color determined by the fixed color determination unit, and a person's face in the first image and its surroundings Image synthesizing means for synthesizing the background portion with a second image whose background is filled with the fixed color.

[0007] Preferably, the fixed color determining means collects color information at a plurality of sampling points set in a background portion around the face of the first image input by the image input means, and calculates the color information of the collected color information. The average value is calculated and used as the fixed color.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing a configuration of a fitting simulation system 10 according to the present invention. 1
Reference numeral 2 denotes a bus, which is shown as one bus without distinguishing between an image bus and other instruction buses. 14 is a CPU for performing various data processing, 16 is a ROM storing various programs, 1
8 is an RA having a read / write work memory
M, 20 are a hard disk, 22 is a frame memory, and 24 is a monitor. Reference numeral 26 denotes an I / O device, which is connected to a capture I / F 28, a video camera 30 serving as image input means described later, a keyboard and mouse 32 for inputting numerical values and commands, and external storage of a drive such as a floppy disk or a magneto-optical disk. The means 34 and the like are connected. Reference numeral 36 denotes a plain back screen when the person 38 is photographed by the video camera 30, and reference numeral 40 denotes illumination used when photographing by the video camera 30.

The RO of the fitting simulation system 10
The M16 and the like are used to form a fixed color determining unit 42 and an image combining unit 44, which will be described later. Similarly, the ROM 16 stores, in addition to these programs, an editing program 46 for performing image editing processing and the like, a program for controlling the video camera 30 and the lighting 40, and the like. Do the job. These means can be realized not only as a program but also as a dedicated IC such as a digital signal processor.

The RAM 18 stores data obtained by processing by the CPU 14 using various means and programs provided in the ROM 16. Here, there is a storage area for each area.
In addition to the area 50 for storing the first image 48 of the person 38 photographed against the background 6, the second image 52 of the model wearing the clothes
The storage area 54 of the first mask data 56 used to paint the background of the second image 52
The storage area 62 of the modified second image 60 obtained by modifying the image 52 by the image combining means 44 using the first mask data 56,
There is a storage area 66 for the second mask data 64 used when the first image 48 and the modified second image 60 are combined.

The first image 48 is a still image stored in the capture I / F 28 by shooting with the video camera 30, and the first image 48 is the first image 48 in the RAM 18.
Is stored in the storage area 50. The second image 52, the first mask data 56, and the second mask data 64 are images corresponding to the selected image by selecting clothes of a desired design by the person 38 performing the fitting simulation from a design selection screen described later. Data and mask data are
Read from the hard disk 20 and the second
The storage area 54 for the image 52, the storage area 58 for the first mask data 56, and the storage area 66 for the second mask data 64 are respectively stored.

Next, the processing of the fitting simulation system 10 of the present invention will be described with reference to FIGS. 2 and 3 to 9.
To do. FIG. 2 is an overall flowchart of the fitting simulation system 10 of the present invention. FIG.
FIG. 4 shows a photographing method of a person 38 performing a fitting simulation, FIG. 4 shows a screen at the time of inputting a face image, FIG. 5 shows a second image 52 of a model wearing clothes, and FIG. 7 is a second image 5 using the first mask data 56.
8 shows a modified second image 60 obtained by modifying the second image, FIG. 8 shows a second mask data 64, and FIG. 9 shows an image 68 obtained by combining a face image in the first image 48 with the second image 52.

First, the process is started in step s1, and in step s2, the person 38 to be subjected to the fitting simulation is photographed using the video camera 30. As shown in FIG. 3, the photographing of the person 38 is performed with a plain back screen 36 as a background. In the present embodiment, the illumination 40 is turned on in synchronization with the shooting by the video camera 30. FIG. 4 shows a screen at the time of inputting a face image of a person. An image taken by the video camera 30 is displayed on the monitor 24. Two horizontal guide lines 70 are displayed on the monitor 24 at the top and bottom, and each guide line 70 has a short vertical line at the center position. Then, in order to match the position of the face to be shot with the upper and lower guidelines 70,
The camera control moves the video camera 30 up and down or left and right, and zooms in and out of the image. Then, when the position adjustment of the video camera 30 is completed, an image is captured. This image is referred to as a first image 48. The first image 48 captured by the video camera 30 is first stored in the capture I / F 28.

In step s3, the fixed color determining means 42 collects color information from a plurality of preset sampling points 72 of the first image 48 stored in the capture I / F 28, and calculates the average value of the color information. calculate. The calculated color information is a fixed color 74. Here, the color information is R
GB, brightness, saturation, and the like.

In the present embodiment, when the face position is adjusted to the guide line 70 as shown in FIG.
Four sampling positions are determined in advance so that 2 becomes the background portion 75 around the face. The number of sampling points 72 is preferably set to be large, but may be small as long as the shooting environment is good. Further, it is preferable to adjust the illumination 40 and the like so that the screen 36 at the sampling location 72 does not become uneven. The sampling location 72
Does not use the one in the preset position,
You may specify it manually.

In step s4, a person 38 who simulates a try-on selects clothes having a desired design. In the hard disk 20, each body type (Y,
A, B, etc.), images of models wearing variously designed clothes are stored, and clothes of a desired design are selected from a plurality of clothes corresponding to one's body displayed on the design selection screen. Select This selected image is referred to as a second image 52. Then, the second image 52 and the first mask data 56 and the second mask data 64 corresponding to the second image 52 are read from the hard disk 20 and are read from the RAM 18.
Are stored in the respective storage areas. In addition, the first image 48 captured by the video camera 30 also has a capture I /
The data is stored in the RAM 18 from F28.

In step s5, the second image of the model wearing desired clothes is determined by the image synthesis means 44 using the first mask data 56 shown in FIG. 6 and the fixed color 74 determined by the fixed color determination means 42. The background 52 and the face portion 76 are painted out. The second image 52 is displayed using the fixed color 74.
The image in which the background portion 76 is filled is the modified second image 60
Is stored in the storage area 62 of the modified second image 60 in the RAM 18. (FIG. 7)

In step s6, the image synthesizing means 44 uses the second mask data 64 shown in FIG. 8 to include the face portion from the first image 48 of the person 38 to be subjected to the fitting simulation, including the background portion around the face. Cut out in state. Modified image of cut-out face and surrounding background portion Second image 60
Combine with At this time, the size and position of the image of the face cut out from the first image 48 and the surrounding background portion are adjusted,
The balance with the size of the model in the modified second image 60 is obtained.

The color of the background portion 78 of the modified second image 60 is the average value of the color information at a plurality of sampling points 72 set in the background portion 75 around the face in the first image 48. There is no discomfort between the color of the background portion 75 around the face extracted together with the face image and the background color of the modified second image 60. In addition, when the image obtained by cutting out the image of the face and its surrounding background portion 75 from the first image 48 using the second mask data 64 is combined with the modified second image 60, semi-transparent synthesis is performed on the surrounding portion. This makes it possible to replace the face image without further discomfort.

By the above processing, an image in which the face of the model in the second image 52 has been replaced with the face of the person 38 for which the fitting simulation is performed can be confirmed on the monitor 24.
It is also possible to print out by connecting a printer. Then, the process proceeds to step s7, and the fitting simulation process ends.

In the present embodiment, the video camera 30 is used as an image input means for inputting the first image 48 of the person 38 to be subjected to the try-on simulation. However, this may be a digital camera or the like. Further, an image of the person 38 to be subjected to the fitting simulation, which is stored in a floppy disk, a magneto-optical disk, or the like, may be input using the external storage unit 34.

Although the preferred embodiment of the present invention has been described,
The present invention is not limited to the above embodiment, and can be implemented without departing from the gist of the present invention.

[0023]

According to the try-on simulation system of the present invention, when the face portion of the person in the first image is replaced with the face of the model wearing the desired clothes in the second image,
The color information of the background of the first image is collected to determine a fixed color, the background of the second image is painted with the determined fixed color, and the person's face in the first image and the background portion around it are fixed with a fixed color. Are synthesized with the filled-in second image. Since image processing by contour extraction is not performed for replacement of the face in the first image and the second image, synthesis of a natural and uncomfortable image is performed.
It can be performed with a simple configuration without using advanced image processing technology. Also, the screen used for photographing the person performing the fitting simulation may be a solid screen, and the color of the screen or the restriction on the environment of the installation location such as lighting can be reduced.

Further, a plurality of sampling points for determining the fixed color are set in the background portion of the first image, and the average value of the color information collected from the plurality of sampling points is calculated to determine the fixed color. When the face is replaced in the image and the second image, it is possible to synthesize a more natural and comfortable image.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a fitting simulation system according to an embodiment of the present invention.

FIG. 2 is a diagram showing an overall flowchart of the fitting simulation system.

FIG. 3 is a diagram illustrating a method of photographing a person performing a fitting simulation.

FIG. 4 is a diagram showing a screen of a monitor when a face image is input.

FIG. 5 is a diagram showing a second image of a model wearing clothes.

FIG. 6 is a diagram showing first mask data.

FIG. 7 is a diagram showing a modified second image obtained by modifying a second image using first mask data.

FIG. 8 is a diagram showing second mask data.

FIG. 9 is a diagram in which a face portion in a first image is combined with a second image.

[Explanation of symbols]

10: fitting simulation system 12: bus 14: CPU 16: ROM 18: RA
M, 20: Hard disk, 24: Monitor, 2
6 ... I / O device, 28 ... Capture I / F, 30
... video camera, 34 ... external storage means, 36 ...
Back screen, 38 ... person, 40 ... lighting,
42: fixed color determining means, 44: image combining means, 4
8 First image, 52 Second image, 56 First mask data, 60 Modified second image, 64 Second mask data, 68 Face image Replaced image, 7
0: guidelines, 72: sampling points, 7
4 ... fixed color, 75 ... background portion around the face of the first image,
76: Background portion of second image

Claims (2)

[Claims] 1. A second image which is a model image of a model wearing clothes stored in a storage means, wherein a person is photographed with a plain screen as a background by an image input means and input as a first image. In a fitting simulation system for combining and replacing a model face of a second image with a human face of a first image captured by the image input means, a fixed color for determining a fixed color by collecting background color information of the first image A color determining means, and the background in the second image is filled with the fixed color determined by the fixed color determining means, and the face of the person in the first image and the background area around the person are converted into a second image in which the background is filled with the fixed color. A try-on simulation system comprising: an image synthesizing means for synthesizing. 2. The fixed color determination means collects color information at a plurality of sampling points set in a background portion around a face of the first image input by the image input means, and calculates an average value of the collected color information. The try-on simulation system according to claim 1, wherein is calculated as a fixed color.