WO2006057267A1 - 顔画像合成方法および顔画像合成装置 - Google Patents
顔画像合成方法および顔画像合成装置 Download PDFInfo
- Publication number
- WO2006057267A1 WO2006057267A1 PCT/JP2005/021517 JP2005021517W WO2006057267A1 WO 2006057267 A1 WO2006057267 A1 WO 2006057267A1 JP 2005021517 W JP2005021517 W JP 2005021517W WO 2006057267 A1 WO2006057267 A1 WO 2006057267A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- face image
- face
- face images
- image
- reference point
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/10—Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
Definitions
- the present invention relates to a face image composition method and a face image composition device that synthesize face images.
- two or more face images, or at least one face image and one face graphic or face animation are synthesized to create a fictitious face image.
- few products have been commercialized.
- feature points in a large number of faces are input in each of the two images as in the three-dimensional polygon mapping 100 shown in Fig. 1, and the faces pasted on the face standard model are synthesized. .
- This method requires extraction of a large number of feature points in the face and complicated calculation, and is difficult to perform on a mobile terminal.
- Japanese Patent Laid-Open No. 2003-242480 discloses an image processing method and a server that can use the method.
- This image processing method matches an existing image stored in the image database of the server with a portrait input to the user terminal, generates an intermediate image in response to the user's customization request, and generates a user terminal.
- the face image is sent to the server via the network, the compositing process is performed at the server, and the result is sent again to the terminal via the network.
- Japanese Laid-Open Patent Publication No. 9-62865 discloses a face image processing method and a face image processing apparatus.
- This face image processing method uses two face images (Gl, G2), a background face image (G3), and each shape model matched to these face images (Gl, G2, G3).
- Paying attention to the characteristic part of the face image (G1) the coordinates of a predetermined point corresponding to the characteristic part of the shape model of the face image (G1) are changed to the corresponding points of the shape model of the background image (G3).
- the shape model and the face image (F1) after matching each of the face images (Gl, G2) Predetermine the coordinate value of each corresponding point in the shape model of G3)
- the shape is complemented with the ratio of the color
- the shape-complemented shape model is color-complemented with a corresponding ratio and a predetermined ratio.
- Japanese Patent Laid-Open No. 10-91808 discloses a face image information conversion method and a face image information conversion apparatus.
- this face image information conversion method the face image information of a plurality of different facial expressions of a predetermined person is processed with respect to each of the facial expressions corresponding to projection components on a plurality of coordinate axes in the face image expression space, with the entire face as a processing unit.
- a step of extracting a plurality of facial expression components; and a plurality of composite facial expression components corresponding to the facial image information of a composite facial expression in an arbitrary ratio between any two specified different facial expressions corresponding to the two different facial expressions Calculating a facial expression component by interpolation; and generating synthetic facial image information in accordance with the plurality of synthetic facial expression components.
- Japanese Patent Application Laid-Open No. 2004-178163 discloses an image processing method and an apparatus therefor.
- the image processing method includes the step of associating the decoration and the arrangement information of the decoration in the body part region; the step of setting the position of the body part region in the input image; and the arrangement information associated with the decoration.
- the step of determining the arrangement of the decoration so as to match the position of the set body part region using and combining the decoration with the input image; and outputting the combined input image is included.
- Japanese Patent Application Laid-Open No. 2004-5265 discloses an image composition method, an image composition device, and an image composition system.
- the image composition method includes a face image acquisition step for acquiring two face images; a base image generation step for generating a base image by combining the two face images by morphing composition; and a feature for acquiring a feature face image.
- JP-A-2003-296713 discloses a face image synthesizing device, a face image synthesizing method, a program for executing the method, a communication terminal including the face image synthesizing device, and a communication method using the communication terminal.
- the face image composition device includes an image input unit that captures an image including a person's face, a face region acquisition unit that acquires a face region including the head of the person from the captured image, and the acquired face Face area processing means for processing image data in the area and creating a new face image.
- Japanese Laid-Open Patent Publication No. 2000-67261 discloses a method and apparatus for creating a face moving image and a recording medium on which this method is recorded.
- This face moving image creation method is a method of creating a moving image by inputting two human face images with different opening / closing states of the mouth and mouth when viewed from the front.
- An eye position detecting step for detecting an eye position from the image; a mouth position detecting step for detecting a mouth position from the two input human face images; and an eye detected in the eye position detecting step.
- An eye deformation control point placement stage in which control points for deformation are arranged around the eyes based on the position information of the eye and the control points of the two human face images are associated with each other; Based on the position information of the mouth detected at the detection stage, control points for deformation are arranged around the mouth, and control points for mouth deformation that associate the control points of the two human face images
- An intermediate image generation stage for generating a plurality of intermediate images by changing the open / closed state of the eyes and mouth by deforming the two human face images inputted based on the point information; A plurality of half-width images generated in the generation step and an image synthesis processing step of synthesizing a human face moving image based on the two input human face images.
- One of the problems of compositing two face images is that if one point is simply weighted and averaged, a place that differs greatly from the color of the skin, such as the eyes, shifts and remains like a ghost, which is unnatural. It is to become a proper synthesis. Therefore, a method using three-dimensional polygon mapping as described above is used. In this method, the above-mentioned conspicuous features can be matched between two faces, so the ghost will not occur. On the other hand, a large amount of feature points in the face are extracted and complicated calculation is performed. In particular, it is difficult to execute on a portable terminal with limited input interface and computing power.
- Facial features include texture and placement of major parts such as eyes, nose and mouth.
- people A's face and person B's face are mixed in a ratio of 7: 3, they must be mixed in two ways. In other words, it is necessary to set the texture mixing ratio to 7 to 3, and the facial parts placement to 7 for person A and 3 for person B.
- the latter requires a complicated calculation when a polygon mapping method is used.
- the facial data is sent to the server via the network and synthesized on the server as disclosed in Japanese Patent Laid-Open No. 2003-242480.
- users who have resistance to reasons such as communication charges and time depending on network conditions.
- An object of the present invention is to provide a face image composition method and a face image composition device capable of performing face image composition in an environment where the input interface and the calculation processing capability are limited, such as a portable terminal. It is.
- Another object of the present invention is to provide a face image synthesizing method and a face image synthesizing apparatus capable of preventing a ghost from being generated in a face synthesized image by shifting a position that differs greatly from the skin color such as eyes. It is to be.
- Still another object of the present invention is to provide a face image synthesizing method and a face image synthesizing apparatus capable of reducing the burden on an operator required for synthesizing face images.
- a face image composition device of the present invention includes a storage device and an arithmetic processing device.
- the storage device stores a plurality of face images and a plurality of feature points including an origin set for each of the plurality of face images.
- the arithmetic processing device generates a plurality of deformed face images obtained by deforming the plurality of face images so that the origins of the plurality of face images coincide with each other and the other plurality of feature points coincide with each other.
- a composite face image is generated from the face image.
- the plurality of feature points include a reference point and a reference point.
- the arithmetic processing unit includes an inter-image scaling unit, an image normalization unit, and an image composition unit.
- the inter-image scaling unit based on the plurality of face images and the plurality of feature points, changes the scale of the plurality of face images so that distances between predetermined feature points coincide with each other. Generate an image.
- the image normalization unit calculates the position of the reference point from the origin in the plurality of scale-changed face images based on the plurality of scale-changed face images and the plurality of feature points.
- the plurality of deformed faces normalized by setting positions of the reference points at the time of image synthesis and moving the reference points of the plurality of scale-changed face images to the positions of the reference points at the time of the face image synthesis.
- the image synthesis unit matches the reference points of the plurality of deformed face images based on the plurality of deformed face images and the plurality of feature points, and at least the color and luminance at a composition ratio in each pixel.
- the composite face image is created by weighting one of them.
- the arithmetic processing unit generates the synthesized face image by averaging the plurality of deformed face images for each pixel.
- the plurality of feature points are a first feature point located in the first quadrant, a second feature point located in the second quadrant, and a third quadrant with respect to the origin. It includes the third feature point located and the fourth feature point located in the fourth quadrant.
- the coordinate of an arbitrary point in an arbitrary face image of the plurality of face images is (xl, yl), and the deformed face image corresponding to the arbitrary point
- the arithmetic processing unit sets the scaling factor to an arbitrary size of the deformed face image. Calculate to be continuous in the region.
- a portable housing including the storage device and the arithmetic processing device, a display unit for displaying the plurality of face images and the composite face image, and the plurality of the plurality of face images. And an input unit capable of inputting feature points.
- the face image synthesis method of the present invention includes: (a) a plurality of face images; and a plurality of feature points including an origin set for each of the plurality of face images. (B) generating a plurality of deformed face images deformed so that positions of two or more feature points included in the plurality of face images match, and (c) the plurality of deformed faces. Generating a composite face image from the face image.
- the plurality of feature points include a reference point and a reference point.
- the scales of the plurality of face images are changed so that distances between predetermined feature points coincide with each other.
- the reference points of the plurality of deformed face images are made coincident and combined with each pixel. Creating a composite face image that is a weighted average of at least one of color and luminance by a ratio.
- the step (b2) includes (b21) the coordinates of an arbitrary point in an arbitrary face image of the plurality of face images is (xl, yl), and the arbitrary image
- the scale factor is the deformed face image.
- the computer program product of the present invention has program code means for executing all the steps described in any one of the above items when used on a computer.
- a computer program product of the present invention has the above program code means stored in a storage means readable by a computer.
- FIG. 1 is an example of a three-dimensional polygon map in the background art.
- FIG. 2 is a configuration diagram of an embodiment of a face image composition device of the present invention.
- FIG. 3 is a configuration diagram of an embodiment of a face image composition device of the present invention.
- FIG. 4 is an external view showing an example of an embodiment of a face image composition device of the present invention.
- FIG. 5 is a flowchart showing an embodiment of a face image synthesis method of the present invention.
- FIG. 6 is a diagram showing a face image synthesis process in the embodiment of the face image synthesis method of the present invention.
- Fig. 7 is a diagram for explaining setting of the enlargement / reduction ratio at the time of face normalization in the embodiment of the face image composition method of the present invention.
- FIG. 2 shows a configuration in the embodiment of the face image composition device of the present invention.
- the face image synthesis device 60 includes an arithmetic processing device 1, a storage device 2, a display device 3, and an input device 4.
- the storage device 2 stores a program 5.
- the arithmetic processing unit 1 is exemplified by a CPU (central processing unit).
- the storage device 2 is exemplified by a memory or a hard disk.
- the display device 3 is exemplified by a liquid crystal display.
- Input device 4 is illustrated on a keyboard.
- FIG. 3 shows a configuration in the embodiment of the face image composition device of the present invention.
- the face image synthesizing device 60 includes coordinate point specifying means 10, inter-image scaling means 11, image normalizing means 12, and image synthesizing means 13.
- the coordinate point designating unit 10 designates a feature point S102 for each of the plurality of input face images S101.
- the feature point S102 includes a reference point (such as under the nose) and a reference point (such as the center point of both eyes and the left and right ends of the mouth).
- the coordinate point specifying means 10 is realized by the display device 3 that displays the input face image S101, the input device 4 that receives the specification of the position on the display screen displayed by the display device 3, and the storage device 2 that stores the coordinates of the specified position. Is done.
- the coordinate point designating means 10 can be omitted if the information specifying the feature point S102 is already obtained when the input face image S101 is input.
- the inter-image scaling unit 11 Based on the feature point S102, the inter-image scaling unit 11 generates an image S103 scaled by the distance between the centers of both eyes among the reference points.
- the image normalization means 12 calculates the position of the reference point when the reference point is the origin, sets the reference point position S106 at the time of composition, and the reference point is obtained from the image S103.
- An image S 104 normalized so as to move to the reference point at the time of synthesis is generated.
- the image compositing means 13 creates a composite face image S 105 by matching the reference points of each normalized face image S 104 and performing weighted average of color or luminance at a desired composition ratio S 107 in each pixel. To do.
- the inter-image scaling unit 11, the image normalization unit 12, and the image composition unit 13 are realized by processing in which the arithmetic processing device 1 reads out the program 5 and executes it according to the procedure described in the program 5.
- FIG. 4 shows an example of the appearance in the embodiment of the face image composition device of the present invention.
- the face image synthesis device 60 is preferably mounted on a mobile terminal device such as a mobile phone. .
- the face image synthesizing device 60 is preferably mounted on a portable terminal device such as a mobile phone having a camera function (not shown).
- the face image synthesis device 60 includes a display device 67.
- the face image synthesizing device 60 includes a pointing device 61 capable of designating an arbitrary position of an image displayed on the display device 67 and a numeric key capable of designating a numerical value.
- the display device 67 corresponds to the display device 3 described above.
- the pointing device 61, the numerical keys 69, and the camera correspond to the input device 4 described above.
- the display device 67 displays the input face image S101.
- the input face image S101 is an image photographed by a camera (not shown) included in the face image synthesizer 60, an image set by default in the storage device 2 included in the face image synthesizer 60, or included in the face image synthesizer 60. An image received by the communication device.
- a reference point 62, an upper left reference point 63, an upper right reference point 64, a lower left reference point 65, and a lower right reference point 66 are set by the pointing device 61.
- the input face image S 101 is a face-like image such as a character without a mouth, only the upper left reference point 63 and the upper right reference point 64 are set as reference points.
- FIG. 5 is a flowchart showing an embodiment of the face image synthesis method of the present invention.
- the flow chart is roughly divided into seven steps, steps S1 to S7.
- the coordinate point designation means 10 performs step S1
- the inter-image scaling means 11 performs step S2
- the image normalization means 12 performs steps S3 to S6, and the image composition means 13 performs step S7.
- FIG. 6 is a diagram showing a face image composition process in the embodiment of the face image composition method of the present invention. An example of how facial images are synthesized is shown.
- Step S1 Two input face images S101 are input by a camera or the like.
- the input face image S101 is an image that can be recognized as a face or face, such as a human face, animal face, doll, or front of a car in the image.
- FIG. 7 shows an example of the input face image S 101 as the face image F1 and the face image F2.
- the coordinate point designation means 10 designates the reference point 62, the upper left reference point 63, the upper right reference point 64, the lower left reference point 65, and the lower right reference point 66 for each input face image S101, and is stored in the storage device 2.
- the This designation may be input by the user using the pointing device 61 without referring to the display device 67, or automatically from the input face image by the program 5 stored in the face image synthesizing device 60. It may be extracted. Standard It is desirable that the points satisfy the conditions such as being easy to extract, having a part different from the skin color nearby, and being intermediate between parts different from the other skin color (both eyes and mouth). An example of a point that satisfies these conditions is the bottom of the nose (the center of the nostril).
- the reference point is a point that indicates the position of the part different from other skin colors such as the center of the right eye, the center of the left eye, and both ends of the mouth, and it is preferable that the part is specified.
- Step S2 The inter-image scaling means 11 generates an image S103 by matching the face scales of the two input face images S101. From the fact that the face sizes included in the input face image S101 are different, even if the image sizes are matched, the face scales are not matched. Therefore, a method of matching the scales using the length of the face part with little individual difference can be considered. Often used is the distance between the eyes. The distance between both eyes is usually around 6cm, regardless of the size of the face or head, as described in the literature on binocular stereopsis.
- the inter-image scaling means 11 holds each input face image S101 horizontally in a line connecting the upper left reference point 63 and the upper right reference point 64 (the face image synthesizer 60 is normally held in an operating posture.
- the display device 67 in the width direction).
- calculate the distance between the centers of both eyes in pixels and calculate the ratio of the distance to the distance between the eyes set in advance, that is, (the distance between the eyes set) / (the distance between the eyes in the image). Multiply the image size by this ratio to enlarge or reduce it.
- the image S103 thus obtained is shown as face images F3 and F4 in FIG. Face image F3 and face image F4 have the same distance between the eyes.
- Step S3 The image normalization means 12 calculates the positional relationship between the main parts (both eyes, nose, mouth) of the face in the image S103. In other words, the position of the reference point when the reference point is the origin is calculated. The pixel coordinate value with the reference point base point as the origin indicates the positional relationship of the major parts of the face.
- Step S4 The image normalization means 12 calculates the positions of the upper left reference point 63, the upper right reference point 64, the lower left reference point 65, and the lower right reference point 66 at the time of synthesis from the two images S103. This is equivalent to determining the position of each reference point with respect to the reference point in the synthesized face image S105.
- a calculation method for example, the following method is used.
- the coordinates of the position of the right eye center of face image A (upper left reference point 63) are (xa, ya) with the reference point as the origin, and the position of the right eye center of face image B Let the coordinates of the position be (xb, yb) and the composition ratio be p: q.
- the position coordinate of the center of the right eye in the composite face image S105 is set by the following equation.
- Step S5 The image normalization means 12 compares the ratio of the distance between the reference point before and after the synthesis of the face image and each reference point (magnification or reduction) for each of the two images S103. Rate, hereinafter abbreviated as scaling rate).
- Rate hereinafter abbreviated as scaling rate.
- Step S6 The image normalization means 12 normalizes (enlarges or reduces) the two images S103. That is, for each of the two images, the reference point is moved to the position after synthesis by multiplying by the ratio calculated in step S5. Further, the points other than the reference point are moved by enlargement or reduction so that the enlargement / reduction ratio calculated in step S5 increases or decreases smoothly according to the position. At this time, enlargement or reduction must be performed so that positions other than the reference point do not become discontinuous. Examples of such enlargement or reduction methods are shown below.
- FIG. 7 is a diagram for explaining setting of the enlargement / reduction ratio at the time of face normalization in the embodiment of the face image composition method of the present invention.
- the face image is divided into 12 parts according to the position with respect to the reference point.
- Regions 31, 32, 33, and 34 in FIG. 7 are regions outside the reference point when viewed from the reference point 20 in terms of both the x coordinate and the y coordinate. In these regions, the enlargement / reduction rate of the innermost point, that is, the reference point is set as the enlargement / reduction rate of an arbitrary point in the region.
- the scaling ratio of region 31 is the scaling ratio of upper left reference point 63.
- the scaling factor for region 32 is the scaling factor for the upper right reference point 64.
- the scaling factor for region 33 is the scaling factor for the lower left reference point 65.
- the enlargement / reduction rate in region 34 is the enlargement / reduction rate at reference point 66 in the lower right.
- one of the x and y coordinates is outside the reference point when viewed from the reference point 20, and the other is within the reference point when viewed from the reference point 20. It is a certain area. These The enlargement / reduction ratio of an arbitrary point located in the region is determined by using the relative position between the point and the reference point in contact with the region. For example, consider the case where the scaling factor of the point (X, y) in the region 41 is obtained. Using the coordinates (xl, yl) of the upper left reference point 63 and the coordinates (X 2, y2) of the upper right reference point 64, the point (X, y)
- Equation 2 The scaling factor of the point (X, y) is expressed by using the scaling factor (rl, r2) of the upper left reference point 63 and the scaling factor (r2, s2) of the upper right reference point 64.
- the coordinates of the upper left reference point 63 and the coordinates of the lower left reference point 65 are used to express the y coordinate of the point inside the area 42 by the internal division relation and set in the same way.
- the coordinates of the upper right reference point 64 and the coordinates of the lower right reference point 66 are used to express the X coordinate of the points inside the region 43 by the internal relation and set in the same way.
- the coordinates of the lower left reference point 65 and the coordinates of the lower right reference point 66 are used to express the y coordinate of the point inside the region 44 by the internal relation and set in the same way.
- Regions 51, 52, 53, and 54 in Fig. 7 are the X-axis and y-axis, where the x-coordinate and y-coordinate are both the reference point forces and the region inside the reference point forms an orthogonal coordinate system with the reference point as the origin. Is divided into four areas. The scaling ratio of these areas 51, 52, 53 and 54 is determined as follows. Description will be made by focusing on the first quadrant area 54 in which both the X coordinate and the y coordinate are positive.
- the coordinates of the point (X, y) in region 54 are the coordinates of the lower right reference point 66 (x4, y4), the lower left reference point 65 (x3, y3), and the upper right reference point 64 (x2, y2) Using
- the image normalization means 12 sets the scaling factor of the points in the region 54 by the following equation.
- each quadrant is enlarged or reduced by the enlargement / reduction ratio of the reference points belonging to each quadrant.
- the eyes and mouth of the input face image S101 are not symmetrical. This is because the continuous line and texture of the input face image S101 are discontinuously deformed in the region crossing the X and y coordinates, resulting in an unnatural face image.
- a natural image in which facial parts, lines, and texture continuously change can be obtained.
- Step S7 The image composition means 13 generates a composite face image S105 using the normalized image S 104 and the input composition ratio S107.
- the synthesized face image S105 is obtained by matching the reference point 62 of the two images S104 normalized in step S6 with the origin, and calculating the luminance value or color value of the pixel at each point on the coordinates by the synthesis ratio S107. Obtained by weighted average. At this time, it is not necessary to synthesize the entire face with a uniform synthesis ratio. For example, two images S104 may be combined only at the circumference of both eyes at a ratio of 50:50, and the other parts may be combined at a ratio of 0: 100 to form a partial composition only around both eyes. Such a composite ratio S 107 is Therefore, it can be stored in the storage device 2 or input device power can be input!
- the input face image S101 is not limited to an image obtained by photographing an actual human face.
- it may be an artificial image created by CG with crab.
- the front of the car the headlights correspond to the eyes
- a person like a doll may be assumed.
- 66 is input using the pointing device 61 while confirming the input face image S101 displayed on the display screen of the display device 67. If a feature point detection program for extracting the input face image S101 force reference point 62 and each reference point is previously installed in the portable terminal, the input by the pointing device 61 is not required.
- Steps S2 and S3 can be executed by a computing device in the mobile terminal.
- Step S4 can be realized, for example, by inputting the composition ratio of the two images S103 with the numeric keys of the portable terminal and performing the above calculation with the arithmetic device in the portable terminal.
- Steps S5 and S6 can be realized by a computing device in the mobile terminal.
- step S7 can be realized by inputting the composite ratio of the two images S104 with the numeric keys of the portable terminal and performing the above-described calculation with an arithmetic device in the portable terminal.
- Toi Reasonability is also preferable. If any of the other four points (for example, the center of the right eye) is used as the reference point, a point away from that point (for example, the left edge of the mouth) is rearranged by enlargement or reduction based on the distance of the reference point force. Shake due to enlargement or reduction tends to increase. If a distance image including three-dimensional shape information can be obtained, the apex of the nose can be used as the reference point.
- the arrangement between the main parts can be defined by the position of the reference point and the remaining four reference points. Therefore, the placement of the reference points after synthesis is determined and the images are enlarged and reduced so that the reference point position matches the post-synthesis position in each synthesized image, so that the reference points match.
- weighted averaging with color or luminance By performing weighted averaging with color or luminance, a composite image free from ghost in the main parts can be obtained, and the object of the present invention can be achieved.
- the first effect of the present invention is that even a limited input interface such as a portable terminal can be executed without burden on the operator.
- face synthesis can be performed by specifying five points: the center of the right eye, the center of the left eye, the right edge of the mouth, and the left edge of the mouth.
- the second effect of the present invention is that it can be executed even with limited calculation processing capability such as a portable terminal. This is because polygon mapping is not used and processing is performed only with enlargement / reduction based on 5 points.
- the third effect of the present invention is to prevent a ghost from being generated in a face composition image, such as an eye, where the skin color greatly differs.
- the reason is that all the original images to be synthesized are enlarged or reduced so that these locations coincide.
Landscapes
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Engineering & Computer Science (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US11/791,626 US7876320B2 (en) | 2004-11-25 | 2005-11-24 | Face image synthesis method and face image synthesis apparatus |
EP05809577A EP1818869A1 (en) | 2004-11-25 | 2005-11-24 | Face image synthesis method and face image synthesis device |
JP2006547806A JP4359784B2 (ja) | 2004-11-25 | 2005-11-24 | 顔画像合成方法および顔画像合成装置 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2004341166 | 2004-11-25 | ||
JP2004-341166 | 2004-11-25 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2006057267A1 true WO2006057267A1 (ja) | 2006-06-01 |
Family
ID=36498003
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2005/021517 WO2006057267A1 (ja) | 2004-11-25 | 2005-11-24 | 顔画像合成方法および顔画像合成装置 |
Country Status (4)
Country | Link |
---|---|
US (1) | US7876320B2 (ja) |
EP (1) | EP1818869A1 (ja) |
JP (1) | JP4359784B2 (ja) |
WO (1) | WO2006057267A1 (ja) |
Families Citing this family (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI332639B (en) * | 2006-09-27 | 2010-11-01 | Compal Electronics Inc | Method for displaying expressional image |
JP2009211513A (ja) * | 2008-03-05 | 2009-09-17 | Toshiba Corp | 画像処理装置及びその方法 |
JP2009232243A (ja) * | 2008-03-24 | 2009-10-08 | Seiko Epson Corp | 画像処理装置、画像処理方法、および画像処理のためのコンピュータプログラム |
EP2353297A2 (en) * | 2008-11-05 | 2011-08-10 | Shachar Carmi | Apparatus and method for chroma-key processing |
JP5496036B2 (ja) * | 2010-09-21 | 2014-05-21 | オリンパス株式会社 | 画像処理装置および画像処理プログラム |
JP5538160B2 (ja) * | 2010-09-24 | 2014-07-02 | パナソニック株式会社 | 瞳孔検出装置及び瞳孔検出方法 |
JP5547854B2 (ja) * | 2010-11-29 | 2014-07-16 | デジタルオプティックス・コーポレイション・ヨーロッパ・リミテッド | 携帯型装置で捕捉される多重画像からの肖像画像合成 |
US9801550B2 (en) * | 2010-12-27 | 2017-10-31 | Joseph Ralph Ferrantelli | Method and system for measuring anatomical dimensions from a digital photograph on a mobile device |
US9788759B2 (en) * | 2010-12-27 | 2017-10-17 | Joseph Ralph Ferrantelli | Method and system for postural analysis and measuring anatomical dimensions from a digital three-dimensional image on a mobile device |
CA2822244C (en) * | 2010-12-27 | 2017-02-28 | Joseph R. Ferrantelli | Mobile postural screening method and system |
US20120182315A1 (en) * | 2011-01-18 | 2012-07-19 | Kenneth Alan Garzo | Make a Face |
JP5820257B2 (ja) * | 2011-02-18 | 2015-11-24 | キヤノン株式会社 | 画像処理装置、撮像装置、画像処理方法、及びプログラム |
US9251402B2 (en) * | 2011-05-13 | 2016-02-02 | Microsoft Technology Licensing, Llc | Association and prediction in facial recognition |
US9323980B2 (en) | 2011-05-13 | 2016-04-26 | Microsoft Technology Licensing, Llc | Pose-robust recognition |
KR102013928B1 (ko) * | 2012-12-28 | 2019-08-23 | 삼성전자주식회사 | 영상 변형 장치 및 그 방법 |
US9811889B2 (en) * | 2014-12-31 | 2017-11-07 | Nokia Technologies Oy | Method, apparatus and computer program product for generating unobstructed object views |
CN104657956B (zh) * | 2015-03-16 | 2018-10-02 | 龙旗电子(惠州)有限公司 | 一种实现智能手机图片美化功能的方法 |
US10460493B2 (en) * | 2015-07-21 | 2019-10-29 | Sony Corporation | Information processing apparatus, information processing method, and program |
US9971958B2 (en) | 2016-06-01 | 2018-05-15 | Mitsubishi Electric Research Laboratories, Inc. | Method and system for generating multimodal digital images |
US11017547B2 (en) | 2018-05-09 | 2021-05-25 | Posture Co., Inc. | Method and system for postural analysis and measuring anatomical dimensions from a digital image using machine learning |
TWI749370B (zh) * | 2019-09-16 | 2021-12-11 | 緯創資通股份有限公司 | 臉部辨識方法及其相關電腦系統 |
JP7102554B2 (ja) * | 2019-09-30 | 2022-07-19 | ベイジン・センスタイム・テクノロジー・デベロップメント・カンパニー・リミテッド | 画像処理方法、装置及び電子機器 |
US11610305B2 (en) | 2019-10-17 | 2023-03-21 | Postureco, Inc. | Method and system for postural analysis and measuring anatomical dimensions from a radiographic image using machine learning |
CN112819922B (zh) * | 2021-02-02 | 2024-03-12 | 郑州轻工业大学 | 一种基于连续线条的人物肖像画生成方法 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0962865A (ja) * | 1995-06-16 | 1997-03-07 | Seiko Epson Corp | 顔画像処理方法および顔画像処理装置 |
JP2000036056A (ja) * | 1998-05-11 | 2000-02-02 | Hitachi Ltd | 画像対応付け方法 |
Family Cites Families (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6463176B1 (en) * | 1994-02-02 | 2002-10-08 | Canon Kabushiki Kaisha | Image recognition/reproduction method and apparatus |
JP2918499B2 (ja) * | 1996-09-17 | 1999-07-12 | 株式会社エイ・ティ・アール人間情報通信研究所 | 顔画像情報変換方法および顔画像情報変換装置 |
JP2000067261A (ja) | 1998-08-19 | 2000-03-03 | Nippon Telegr & Teleph Corp <Ntt> | 顔動画像作成方法及び装置及びこの方法を記録した記録媒体 |
JP2000357221A (ja) * | 1999-06-15 | 2000-12-26 | Minolta Co Ltd | 画像処理装置および画像処理方法、ならびに画像処理プログラムを記録した記録媒体 |
US6504546B1 (en) * | 2000-02-08 | 2003-01-07 | At&T Corp. | Method of modeling objects to synthesize three-dimensional, photo-realistic animations |
US6975750B2 (en) * | 2000-12-01 | 2005-12-13 | Microsoft Corp. | System and method for face recognition using synthesized training images |
US6654018B1 (en) * | 2001-03-29 | 2003-11-25 | At&T Corp. | Audio-visual selection process for the synthesis of photo-realistic talking-head animations |
JP2003242480A (ja) | 2002-02-14 | 2003-08-29 | Sony Communication Network Corp | 画像処理方法およびその方法を利用可能なサーバ |
JP2003271982A (ja) * | 2002-03-19 | 2003-09-26 | Victor Co Of Japan Ltd | 似顔絵生成装置 |
JP2003296713A (ja) | 2002-04-04 | 2003-10-17 | Mitsubishi Electric Corp | 顔画像合成装置、顔画像合成方法およびその方法を実行するプログラムならびに顔画像合成装置を備えた通信用端末およびその通信用端末による通信方法 |
JP2004005265A (ja) | 2002-05-31 | 2004-01-08 | Omron Corp | 画像合成方法、画像合成装置、画像合成システム |
JP2004178163A (ja) | 2002-11-26 | 2004-06-24 | Matsushita Electric Ind Co Ltd | 画像処理方法及びその装置 |
US7587068B1 (en) * | 2004-01-22 | 2009-09-08 | Fotonation Vision Limited | Classification database for consumer digital images |
US7317815B2 (en) * | 2003-06-26 | 2008-01-08 | Fotonation Vision Limited | Digital image processing composition using face detection information |
US7379071B2 (en) * | 2003-10-14 | 2008-05-27 | Microsoft Corporation | Geometry-driven feature point-based image synthesis |
-
2005
- 2005-11-24 EP EP05809577A patent/EP1818869A1/en not_active Withdrawn
- 2005-11-24 JP JP2006547806A patent/JP4359784B2/ja active Active
- 2005-11-24 WO PCT/JP2005/021517 patent/WO2006057267A1/ja active Application Filing
- 2005-11-24 US US11/791,626 patent/US7876320B2/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0962865A (ja) * | 1995-06-16 | 1997-03-07 | Seiko Epson Corp | 顔画像処理方法および顔画像処理装置 |
JP2000036056A (ja) * | 1998-05-11 | 2000-02-02 | Hitachi Ltd | 画像対応付け方法 |
Also Published As
Publication number | Publication date |
---|---|
JP4359784B2 (ja) | 2009-11-04 |
EP1818869A1 (en) | 2007-08-15 |
JPWO2006057267A1 (ja) | 2008-08-07 |
US20080165187A1 (en) | 2008-07-10 |
US7876320B2 (en) | 2011-01-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4359784B2 (ja) | 顔画像合成方法および顔画像合成装置 | |
JP3847753B2 (ja) | 画像処理装置、画像処理方法、記録媒体、コンピュータプログラム、半導体デバイス | |
JP7504968B2 (ja) | アバター表示装置、アバター生成装置及びプログラム | |
CN112288665B (zh) | 图像融合的方法、装置、存储介质及电子设备 | |
EP1710746A1 (en) | Makeup simulation program, makeup simulation device, and makeup simulation method | |
JP5632469B2 (ja) | キャラクタ生成システム、キャラクタ生成方法及びプログラム | |
KR101150097B1 (ko) | 얼굴 화상 작성장치 및 방법 | |
CN107452049B (zh) | 一种三维头部建模方法及装置 | |
CN109949390A (zh) | 图像生成方法、动态表情图像生成方法及装置 | |
TWI780919B (zh) | 人臉影像的處理方法、裝置、電子設備及儲存媒體 | |
US20100202697A1 (en) | Specifying position of characteristic portion of face image | |
US8643679B2 (en) | Storage medium storing image conversion program and image conversion apparatus | |
JP4468631B2 (ja) | 3次元顔モデルのためのテクスチャー生成方法及び装置 | |
JP4689548B2 (ja) | 画像処理装置、画像処理方法、記録媒体、コンピュータプログラム、半導体デバイス | |
JP6402301B2 (ja) | 視線変換装置、視線変換方法及びプログラム | |
JP5642583B2 (ja) | 画像生成装置および画像生成プログラム | |
JP2943703B2 (ja) | 3次元キャラクタ作成装置および3次元キャラクタ作成方法 | |
US6633291B1 (en) | Method and apparatus for displaying an image | |
JP2723070B2 (ja) | 人物像表示によるユーザインタフェース装置 | |
JP3543152B2 (ja) | 画像作成装置および画像作成方法 | |
JP2000293710A (ja) | 三次元似顔絵作成方法ならびに装置 | |
JP2001307122A (ja) | 顔写真画像切り抜き方法 | |
JP2000353252A (ja) | 映像重畳方法、映像重畳装置及び映像重畳プログラムを記録した記録媒体 | |
JP4767331B2 (ja) | 画像処理装置、画像処理方法、記録媒体、コンピュータプログラム、半導体デバイス | |
JP2001216525A (ja) | 画像処理装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KM KN KP KR KZ LC LK LR LS LT LU LV LY MA MD MG MK MN MW MX MZ NA NG NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SM SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): BW GH GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LT LU LV MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DPE1 | Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101) | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2006547806 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2005809577 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 11791626 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWP | Wipo information: published in national office |
Ref document number: 2005809577 Country of ref document: EP |