JP2004078637A - Facial portion extracting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To resolve a problem in a conventional facial portion extracting device by template matching wherein a portable device is not realized since a calculation cost is high and miniaturization and reduction of electric power consumption are difficult. <P>SOLUTION: Extraction by shapes of portions is not carried out, and portion extraction focused on edge components in a facial area is carried out. An edge image is created from an input image, portion searching ranges for eyes and mouth are determined in consideration of a position and size of the facial area and geometrical characteristics of a face, and extraction of each portion is carried out by a technique with a low calculation cost. Since a template image is not stored, memory usage is small. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a face part extracting apparatus for extracting an eye and a mouth included in a face in an image.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, a face part extracting apparatus for extracting an eye and a mouth region included in a face in an image has been disclosed in Japanese Patent Application Laid-Open No. 8-287216, for extracting a part such as an eye and a mouth inside a face region. In addition, the position of an eye or a mouth or the like is determined by matching an input image with a template image of a part such as an eye or a mouth prepared in advance in the face region.
[0003]
[Problems to be solved by the invention]
However, in the above-described conventional apparatus, the calculation cost is large for performing the template matching, and it is difficult for a device having relatively low processing performance, such as a portable device, to perform the process in a time without stressing the user.
[0004]
In addition, in order to determine the position of the mouth with respect to both the image with the mouth closed and the image with the mouth slightly opened, a plurality of template images of the mouth are prepared and a plurality of template images are prepared for the input image. There is a problem that it is necessary to perform matching with the image, a large amount of memory is used to hold the template image, and a long time is required for processing.
[0005]
[Means for Solving the Problems]
In order to solve the above problem, the present invention does not perform computationally expensive processing such as template matching, takes into account the position and size of the face region and the geometric feature of the face, and performs the following processing on the edge information. It is characterized in that the eye and mouth parts are extracted by performing such an analysis having a relatively low calculation cost.
[0006]
In order to determine the vertical range of both eyes with a small calculation cost, an edge image is created, the width is the width of the face area, and the height is the edge within the height range determined by multiplying the face height by a constant. A plurality of component sums are calculated within a range where both eyes can exist in the face region. From the plurality of calculated edge total values, a candidate representing the region of the most eye is selected, and the range is determined as the range of both eyes.
[0007]
After determining the vertical range of both eyes with the above method, in order to determine the more accurate vertical range of the right and left eyes, expand the range of the vertical range of both eyes determined by the above method up and down again by the same method. , Right and left eyes.
[0008]
At this time, the width of the area for calculating the total edge, which is the width of the face area in the above method, is set to about half of the width, and the height is the same as the above method. Since the eyes are biased to the right and left in the face area, respectively, the search range of the eyes is set to the right and left sides in the face area, and the edge sum is calculated.
[0009]
In determining the vertical range of the one eye, the width of the search range of the right and left eyes is larger than half the width of the face area, and by giving an overlap, even when the person to be imaged is not facing directly in front, Ensure that the vertical range can be determined correctly.
[0010]
The horizontal range of the eyes is determined by multiplying the sum of the edge components within the range determined by multiplying the width of the face area by a constant and having the height of the vertical range of the eyes determined above. Plural calculations are performed inside the right and left halves of the face within the vertical range. From the plurality of calculated edge total values, a candidate representing the region of the eye is selected, and the range is determined as the horizontal range of the eye.
[0011]
Next, in order to determine the vertical position between the upper lip and the lower lip with a small calculation cost, the sum of the edge components of each line within the range where the mouth can exist in the face region is calculated. From the plurality of calculated edge total values, a candidate representing the region between the upper lip and the lower lip is selected, and the line is determined to be between the upper lip and the lower lip.
[0012]
The vertical range where the mouth can exist is below the eye region in the face region, and the horizontal range is between the centers of the right and left eyes.
[0013]
A threshold value is obtained from the sum of the edges of the line determined to be between the upper lip and the lower lip, and the threshold value is compared with the sum of the edges of each line in order from the line determined to be between the upper lip and the lower lip.
[0014]
Since the edge component at the boundary between the lip and the skin is larger than the edge component at the boundary between the upper lip and the lower lip, a portion exceeding the threshold is set as the upper end of the mouth.
[0015]
The determination of the lower end is performed in the same manner, thereby determining the vertical range of the mouth. Regarding the horizontal range, the region of the mouth is determined by setting the range between the centers of the right and left eyes.
[0016]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 1 is a diagram showing a functional configuration of an embodiment of a face part extracting apparatus according to the present invention.
[0017]
Hereinafter, the configuration and operation of the face part extracting apparatus will be described mainly with reference to FIG.
[0018]
In FIG. 1, reference numeral 1 denotes an input image for extracting a face part, and 2 denotes a face area extracting unit for extracting a face area in the input image 1. Reference numeral 3 denotes an edge image creation unit that creates an image in which an edge component has been calculated from an input image. Reference numeral 4 denotes an eye extraction unit for extracting right and left eye regions, and reference numeral 5 denotes a mouth extraction unit for extracting a mouth region. Reference numeral 21 denotes a user correction input unit for inputting a correction to the extracted eye region from the user.
[0019]
When the input image 1 is input to the face area extraction unit 2 in the face part extraction device of FIG. 1, a face area in the input image 1 is extracted. The face area can be extracted by, for example, performing color space conversion on the input image 1. The input image is converted into the HSV space, threshold processing is performed on the H (Hue: hue) component, and the flesh color portion of the person is determined. The face area in the input image 1 is determined in consideration of the size and shape of the determined skin color area.
[0020]
Normally, the input image 1 is captured as if a human is looking at the camera in front, and an image as shown in FIG. 2A is obtained.
[0021]
In FIG. 2, 22 is a head, 23 and 24 are right and left eyes, 25 is a nose, and 26 is a mouth.
[0022]
When the face area is extracted by the face area extraction unit 2, a face area 27 is obtained as shown in FIG. The face area information extracted by the face area extracting unit 2 is transmitted to the eye extracting unit 4 and the mouth extracting unit 5 and is referred to when extracting the eyes and the mouth.
[0023]
The edge image creation unit 3 extracts an edge component from the input image 1 and creates an edge image. The created edge image is transmitted to the eye extracting unit 4 and the mouth extracting unit 5 and is referred to when extracting the eyes and the mouth. Since the eyes and the mouth include more lines extending in the horizontal direction than in the vertical direction, the edge image creating unit 3 creates an edge image using a filter sensitive to the edge of the horizontal line segment. The eyes and mouth can be stably extracted in the mouth extraction unit 5. When extracting the parts of the eyes and the mouth, only the edge information inside the face area is used, so the face area information extracted by the face area extraction unit 2 is transmitted to the edge image creation unit 3 and only the inside of the face area is extracted. May be created.
[0024]
First, in order to determine the vertical range of both eyes from the image created by the edge image creating unit 3, processing is performed according to the procedures shown in 6 to 8 in FIG.
[0025]
When the face area extraction unit 2 extracts the face area 27 as shown in FIG. 3A, both eyes are present above the face area 27. Therefore, in the edge image created by the edge image creating unit 3 as shown in FIG. 3B, the upper part in the face area 27 is set as the binocular search area 28.
[0026]
The upper left coordinates of the face area 27 are (fx1, fy1), the lower right coordinates are (fx2, fy2), the upper left coordinates of the binocular search area 28 are (sex1, sey1), and the lower right coordinates are (sex2, sey2). ).
[0027]
At this time, sex1 = fx1, sey1 = fy1, sex2 = fx2, and sey2 = fy1 + a × (fy2-fy1).
[0028]
a takes a value from 0 to 1, and the larger the a is, the wider the binocular search area 28 is.
[0029]
The eye area above the face area 27 is where the edge component is localized. In order to extract the vertical range of both eyes at a low calculation cost, the width is the same as that of the binocular search area 28 and the height of the face area 27 is within the binocular search area 28 as shown by 29 in FIG. An edge total calculation range 29 having a height determined by multiplying the height by a constant is determined ("binocular search range and height determination processing for edge total calculation" step 6 in FIG. 1).
[0030]
Next, the edge sum in the edge sum calculation range 29 is calculated at a plurality of positions in the binocular search area (step 7 of “processing for calculating the edge sum at each position in the search range” in FIG. 1).
[0031]
From the plurality of calculated edge total values, a candidate representing the region of the most eye (for example, the one with the largest edge total) is selected, and as shown in FIG. (“Vertical range determination process for both eyes” in FIG. 1, step 8).
[0032]
In order to determine the vertical range of both eyes, it is only necessary to calculate and compare a plurality of edge sums, so that the calculation cost is low.
[0033]
In addition, when the set range 29 includes a part where both eyes are present, the edge components are distributed evenly within the set range 29 as compared with the case where an edge part such as a boundary between skin and hair is included. As a result, the total value of the edge components becomes extremely high as compared with other places, so that the vertical range of both eyes can be determined stably.
[0034]
In addition, if an attempt is made to determine the vertical position of both eyes only by the sum of the edges of one line, erroneous extraction may occur due to the influence of the edge of the input image noise. By calculating, the influence of noise can be reduced, and stable extraction can be performed.
[0035]
In addition, the height of the range 29 is such that the distance between the upper eyelid and the lower eyelid is set so that both eyes fall within the range and include no other part as much as possible so that the vertical range of both eyes can be determined stably. Eye level).
[0036]
Assuming that the upper right coordinate of the range 29 is (sex3, sey3) and the lower left coordinate is (sex4, sey4), it can be expressed as sex3 = sex1, sex4 = sex2, and sey4 = sey3 + b × (fy2-fy1).
[0037]
Note that b is set to a constant such that b × (fy2−fy1) is about the height of the eyes. Although the distance between the upper and lower eyelids (eye height) is slightly different between individuals, it is possible to calculate an appropriate value by multiplying the face height by a constant.
[0038]
Next, in order to determine the vertical range of one eye with reference to the vertical range 30 of both eyes determined as described above, processing is performed according to the procedures 9 to 11 in FIG.
[0039]
When the vertical range 30 of both eyes is determined as shown in FIG. 4A, the search range of one eye is extended up and down from the vertical range 30 as shown in FIGS. One-eye search ranges 31 and 32 in a range from the right and left ends of the region 27 to the vicinity of the center are determined ("Right and left eye search ranges and height determination processing for edge total calculation" step 9 in FIG. 1), and With respect to them, processing is performed in the same procedure as the determination of the vertical range of both eyes.
[0040]
The coordinates of the upper right of the range 30 of both eyes are (ex1, ey1), the coordinates of the lower left are (ex2, ey2), the coordinates of the upper right of the search range 31 are (sex5, sey5), and the coordinates of the lower left are (sex6, sey6). ), Sex5 = ex1, sey5 = ey1-c * (fy2-fy1), sex6 = (1-e) * ex1 + e * ex2, and sey6 = ey2 + d * (fy2-fy1).
[0041]
Here, c and d are parameters indicating how much the search range is expanded upward and downward.
[0042]
E is a parameter that takes a value from 0 to 1 and determines the width of the search area.
[0043]
As shown in FIGS. 4D and 4E, a calculation range 33 having the same width in the horizontal direction as the search range 31 and a height in the vertical direction determined by multiplying the height of the face area 27 by a constant. The internal edge sum is calculated at a plurality of positions ("edge sum calculation process at each position within search range" in FIG. 1) (step 10).
[0044]
Assuming that the upper right coordinate of the range 33 is (sex7, sey7) and the lower left coordinate is (sex8, sey8), it can be expressed as sex7 = sex5, sex8 = sex6, and sey8 = sey7 + b × (fy2-fy1).
[0045]
From the calculated plurality of edge total values, the candidate representing the eye region is selected and determined as the vertical range 34 of one eye ("vertical range determination processing for right and left eyes" step 11 in FIG. 1).
[0046]
The other eye is processed in the same manner to determine the vertical range 35.
[0047]
At this time, the search range 31 of one eye (right eye) and the search range 32 of the other eye (left eye) overlap each other near the center of the face as shown in FIGS. Can be dealt with in a right or left direction instead of directly in front.
[0048]
On the other hand, if there is no overlapping portion between the search range 31 of one eye and the search range 32 of the other eye, as shown in FIGS. And cannot be extracted correctly.
[0049]
The horizontal range is determined for each eye determined as described above according to the procedure shown in 12 to 14 in FIG.
[0050]
When the vertical range of one eye is determined as shown in FIG. 6A in the above processing, the horizontal range of the search range used in determining the vertical range is the vertical range determined as shown in FIG. Are set as search ranges 34 and 35 of one eye (“Right and left eye search ranges and edge total calculation width determination process” procedure 12 in FIG. 1).
[0051]
As shown in FIGS. 6B and 6C, the vertical direction is the same height as the search range 34, and the horizontal width is the edge within the range 38 having a width determined by multiplying the width of the face area 27 by a constant. The sum of the values is calculated at a plurality of positions ("edge total calculation process at each position within the search range" in FIG. 1) (step 13).
[0052]
Note that the upper right coordinate of the vertical range 34 of one eye is (ex3, ey3), the lower left coordinate (ex4, ey4), the upper right coordinate of the right eye edge total calculation range 38 is (sex9, sey9), and the lower left coordinate is (ex9, sey9). Assuming that sex10, sey10), it can be expressed as sey9 = ey3, sex10 = sex9 + f × (fx2-fx1), and sey10 = ey4 = sey9 + b × (fy2-fy1).
[0053]
f is set so that fx (fx2-fx1) is about the width of the eye.
[0054]
Although the eye width is slightly different between individuals, it is possible to calculate an appropriate value by multiplying the face width by a constant.
[0055]
From the plurality of calculated edge total values, a candidate representing the region of the most eye is selected and determined as the horizontal range 39 of one eye (the extracted right eye) (“horizontal range determination processing of right and left eyes” in FIG. 1) 14). The other eye (left eye) is processed in the same manner, and the horizontal range 40 is determined.
[0056]
Next, the detection of the center of the mouth (the boundary between the upper lip and the lower lip) is performed according to the procedures shown in FIGS.
[0057]
The height and width of the face area 27 are multiplied by constants, and the horizontal position is set near the center of the face area 27, and the vertical position is set below the face area 27 as the mouth center search range 41 as shown in FIG. Step 15 of "determination processing of search range in center of mouth" in FIG.
[0058]
Assuming that the upper left coordinate of the mouth center search range 41 is (smx1, smy1) and the lower left coordinate is (smx2, smy2), it is expressed as smx2 = smx1 + g × (fx2-fx1) and smy2 = smy1 + h × (fy2-fy1). Can be.
[0059]
g is set so that g × (fx2−fx1) is about the width of the mouth.
[0060]
Although the width of the mouth slightly differs between individuals, it is possible to calculate an appropriate value by multiplying the width of the face by a constant.
[0061]
Also, h is selected to be a value that is about half the height of the face area 27.
[0062]
smx1 is a value such that the lower end of the mouth center search range 41 does not protrude from the lower end of the face region 27 so that the mouth center search range 41 is located at the center of the face area 27.
[0063]
Then, a plurality of edge value sums of the pixel line 42 having the same width as the mouth center search range 41 are calculated in the mouth center search range 41 (“the edge sum calculation process at each position within the search range” in FIG. 1, step 16). ).
[0064]
From the plurality of calculated edge total values, a candidate representing the center of the mouth (for example, the one with the largest edge total) is selected, and the position is determined as the mouth center position 43 as shown in FIG. (Step 17 of "vertical position determination process of the center of the mouth" in FIG. 1).
[0065]
When determining the mouth center search range 41, the mouth is usually located in the face area and below the position of the eyes, and is located in the area between the perpendiculars lowered from the centers of the right and left eyes, respectively. The determination may be made using information of the face region 27 and the positions of the eyes (horizontal range) 39 and 40.
[0066]
At this time, it can be expressed as smx1 = (ex5 + ex6) １2, smy1 = max (ey6, eye8) + i, smx2 = (ex7 + ex8) ｍ2, and smy2 = fy2-j.
[0067]
However, in the captured image, the smaller horizontal coordinate in the left eye extraction range is ex5 and the larger horizontal coordinate is ex6. Similarly, the smaller horizontal coordinate in the right eye extraction range is ex7 and the larger horizontal coordinate is ex7. The coordinates are ex8.
[0068]
Also, i is a parameter that sets how much the mouth is extracted from the eye extraction range, and is preferably a value that is about 1 to 1.5 times the eye height.
[0069]
In addition, j is a parameter for setting how far above the lower end of the face area 27 the mouth is to be extracted, and is preferably about the eye level.
[0070]
Note that max () is a function for calculating the maximum value from the arguments. By setting the width of the mouth center search range 41 to be the width between the centers of the right and left eyes, the horizontal range of the mouth is determined.
[0071]
Next, in order to determine the positions of the upper end and the lower end of the mouth, the processing is performed according to the procedures shown in 18 to 20 in FIG.
[0072]
A value obtained by multiplying the sum of the edge values at the mouth center position 43 by a constant is calculated as a threshold value (“Threshold determination process” step 18 in FIG. 1).
[0073]
Next, as shown in FIG. 8B, in order to determine the lower end of the mouth, the total edge value in the mouth center search range 41 is calculated one pixel line downward from the mouth center position 43, and the pixel line is determined. The sum of the edge values of No. 44 is compared with the threshold value (Step 19 of “calculation of the sum of edges at each position within the search range and comparison with the threshold value” step 19).
[0074]
Based on the result of the comparison (for example, when the sum of the edge values is larger than the threshold value), the position is determined as the lower end 44 of the mouth as shown in FIG.
[0075]
Similarly, the upper end 45 of the mouth is determined, and the upper end 44 and the lower end 45 of the mouth are determined ("vertical position determination process of upper and lower ends of the mouth" step 20 in FIG. 1).
[0076]
The region of the mouth is determined by the determination of the horizontal range of the mouth and the upper and lower ends of the mouth.
[0077]
In addition, when the face part extracting apparatus is provided with a user correction input means 21 (for example, a keyboard) of FIG. 1 by an input means from a user, an extraction result by an eye extraction unit 4 shown in FIG. If the result is not good, the user adjusts the extraction position using the correction input unit 21.
[0078]
As a result, the extracted eye ranges 39 and 40 cannot be correctly extracted as shown in FIG. 9A, and the mouth determined by the incorrect eye extraction positions 47 and 48 as shown in FIG. 9C. The search range (rather than the range of the mouth center search range 41) becomes an undesired mouth search range 49, and the range of the extracted mouth (rather than the extracted mouth range 46) is correct as shown in FIG. It is possible to prevent a shift as in the range 50 of no mouth, and it is possible to more accurately extract the mouth.
[0079]
As described above, the edge image is created only once for the input image, and the calculation cost for the extraction processing of each part is not high, so that a portable and compact device can be realized.
[0080]
【The invention's effect】
As described above, according to the present invention, the vertical range of both eyes can be determined with a small calculation cost while minimizing the influence of noise included in an input image.
[0081]
Further, according to the present invention, the vertical range of the right and left eyes can be determined even when the captured face image is inclined. In addition, since the search is performed from the surroundings based on the vertical range of both eyes, the calculation cost required for the search is small, and a stable range can be determined.
[0082]
Further, according to the present invention, the vertical range of the right and left eyes can be determined stably even when the captured face image does not face the front.
[0083]
Also, according to the present invention, the determination of the horizontal range of the right and left eyes can be realized with a small calculation cost.
[0084]
Further, according to the present invention, a stable center of the mouth can be extracted with a small calculation cost by narrowing the search range.
[0085]
Also, according to the present invention, it is possible to set the mouth search range more accurately and in a narrower range than when only the face area information is used.
[0086]
Further, according to the present invention, it is possible to calculate an appropriate threshold value in accordance with an input image and to extract the upper and lower ends of the mouth even when extraction is difficult by pattern matching such as when the mouth is open.
[0087]
Furthermore, according to the present invention, highly accurate extraction of the mouth position can be performed interactively by inputting the correction of the extraction information from the user.
[Brief description of the drawings]
FIG. 1 is a diagram showing a functional configuration showing one embodiment of the present invention.
FIG. 2 is a diagram showing an image of a person and an extracted face area.
FIG. 3 is a diagram showing a procedure for extracting a vertical range of both eyes.
FIG. 4 is a diagram showing a procedure for extracting a vertical range of right and left eyes.
FIG. 5 is a diagram showing the difference between the case where the search range in the horizontal direction is overlapped and the case where the search range in the horizontal direction is not set for extraction of the right and left eyes when the person is not facing the front;
FIG. 6 is a diagram showing a procedure for extracting a horizontal range of right and left eyes.
FIG. 7 is a diagram showing an extraction procedure between an upper lip and a lower lip.
FIG. 8 is a diagram showing an extraction procedure of an upper end and a lower end of a mouth.
FIG. 9 is a diagram showing the result of mouth extraction when the eye extraction result is correct and when it is not.
[Explanation of symbols]
Reference Signs List 1 input image 2 face area extraction unit 3 edge image creation unit 4 eye extraction unit 5 mouth extraction unit 6 binocular search range and height determination processing for edge total calculation 7 processing for calculating edge total at each position in search range 8 Vertical range determination processing for both eyes 9 Search range for right and left eyes and height determination processing for edge total calculation 10 Edge total calculation processing at each position within the search range 11 Vertical range determination processing for right and left eyes 12 Right and left eyes Search range and edge total calculation width determination processing 13 Edge total calculation processing at each position within the search range 14 Horizontal range determination processing for the right and left eyes 15 Search area determination processing at the center of the mouth 16 Edge at each position within the search range Total calculation processing 17 Vertical position determination processing at the center of the mouth 18 Threshold determination processing 19 Calculation of the sum of the edges at each position within the search range and comparison processing with the threshold 20 Vertical processing at the upper and lower ends of the mouth Placement determination processing 21 User correction input unit 22 Human head 23 Right eye 24 Left eye 25 Nose 26 Mouth 27 Face area 28 Binocular search area 29 Edge total calculation range 30 Extracted binocular range 31 Right eye search range 32 Left eye search range 33 Calculation range 34 Vertical range 35 Vertical range 36 Search range 37 Search range 38 Right eye edge total calculation range 39 Horizontal range 40 Horizontal range 41 Center of mouth search range 42 Pixel line 43 Center of mouth 44 Pixel line (bottom of mouth)
45 pixel lines (top of mouth)
46 extracted mouth range 47 incorrect right eye range 48 incorrect left eye range 49 undesirable mouth search range 50 incorrect mouth range

Claims (8)

A face area extracting means for extracting a face area from the input image;
Edge component extraction means for extracting edge components from the input image;
Edge component sum calculating means for calculating the sum of the edge components within an arbitrary range, the sum of the edge components within a range of height and width determined from the height and width of the face region, Calculated at multiple positions within the determined binocular search range,
A face part extracting apparatus characterized in that the vertical position of both eyes is determined by comparing the calculated sum of a plurality of edge components. A vertical range in which the sum of the edge components within the range of height and width determined by the height and width of the face area is extended vertically from the determined vertical position of both eyes by the distance determined by the height of the face area , Calculated at a plurality of positions in a horizontal range having a width determined from the width of the face region biased to the right or left in the face region,
2. The face part extracting apparatus according to claim 1, wherein the vertical positions of the right and left eyes are determined by comparing the calculated sum of the plurality of edge components. 3. The face part extracting apparatus according to claim 2, wherein the horizontal range deviated rightward and the horizontal range deviated leftward have a part overlapping. The sum of the edge components within the range of height and width determined from the height and width of the face area is calculated at a plurality of horizontal positions at the vertical positions of the right and left eyes,
The face part extracting apparatus according to claim 1, wherein the horizontal positions of the right and left eyes are determined by comparing the calculated sum of the plurality of edge components. A face area extracting means for extracting a face area from the input image;
Edge component extraction means for extracting edge components from the input image;
Edge component sum calculating means for calculating the sum of the edge components within an arbitrary range, wherein the sum of the edge components at each pixel line having a width determined by the width of the face region is calculated based on the height and width of the face region. Calculated at multiple positions within the determined mouth extraction range,
A face part extracting apparatus characterized in that the vertical position of a mouth is determined by comparing the calculated sum of a plurality of edge components. 6. The face part extracting apparatus according to claim 5, wherein the mouth extraction range is determined not only from the face area but also from the position information of the right and left eyes. Determine a threshold from the sum of the edge components at the determined vertical position of the mouth,
Within the mouth extraction range determined from the face area, sequentially scan each pixel line above and below the determined vertical position of the mouth, compare with the determined threshold, and determine the upper and lower ends of the mouth area. The face part extracting apparatus according to claim 5, wherein: 7. The face part extracting apparatus according to claim 6, wherein the extraction range of the mouth changes according to the user's adjustment of the eye extraction result.

Images