WO2020181900A1 - 图像处理方法及装置、图像设备及存储介质 - Google Patents

图像处理方法及装置、图像设备及存储介质 Download PDF

Info

Publication number
WO2020181900A1
WO2020181900A1 PCT/CN2019/130970 CN2019130970W WO2020181900A1 WO 2020181900 A1 WO2020181900 A1 WO 2020181900A1 CN 2019130970 W CN2019130970 W CN 2019130970W WO 2020181900 A1 WO2020181900 A1 WO 2020181900A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
vector
adjusted
sub
region
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/130970
Other languages
English (en)
French (fr)
Inventor
李通
刘文韬
钱晨
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Sensetime Technology Development Co Ltd
Original Assignee
Beijing Sensetime Technology Development Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Sensetime Technology Development Co Ltd filed Critical Beijing Sensetime Technology Development Co Ltd
Priority to JP2020558530A priority Critical patent/JP7109585B2/ja
Priority to SG11202010399VA priority patent/SG11202010399VA/en
Publication of WO2020181900A1 publication Critical patent/WO2020181900A1/zh
Priority to US17/073,769 priority patent/US11538207B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—Two-dimensional [2D] image generation
    • G06T11/60—Creating or editing images; Combining images with text
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T13/00—Animation
    • G06T13/20—Three-dimensional [3D] animation
    • G06T13/40—Three-dimensional [3D] animation of characters, e.g. humans, animals or virtual beings
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G06T7/0014—Biomedical image inspection using an image reference approach
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/24—Aligning, centring, orientation detection or correction of the image
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
    • G06V10/469—Contour-based spatial representations, e.g. vector-coding
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G06V10/754—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries involving a deformation of the sample pattern or of the reference pattern; Elastic matching
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/60—Type of objects
    • G06V20/64—Three-dimensional [3D] objects
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/174—Facial expression recognition
    • G06V40/176—Dynamic expression
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2200/00—Indexing scheme for image data processing or generation, in general
    • G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person
    • G06T2207/30201—Face

Definitions

  • the present disclosure relates to the field of information technology, and in particular to an image processing method, device, equipment, and storage medium.
  • the deformation is generally based on the user's manual operation, for example, image processing software such as photoshop, which largely depends on the user's operating skills; for ordinary users, the operation is difficult Big.
  • image processing software such as photoshop
  • Another type of image processing software appears in the related art. The user performs a one-click operation, and the image processing software processes the image as a whole.
  • this type of operation performed by electronic equipment is limited to the mechanical nature of the equipment processing. (That is, lack of intelligence), the processed image effect is a bit strange, and the image processing effect is not as expected.
  • the embodiments of the present disclosure expect to provide an image processing method, device, device, and storage medium.
  • an image processing method including:
  • the area to be adjusted is the same or different from the reference area.
  • the determining the orientation of the reference area according to the key points of the reference area includes:
  • the at least three key points include a first key point, a second key point, and a third key point; the first key point and the third key point are related to the second key point.
  • Key points are symmetrical;
  • the determining the target vector based on the at least three key points includes:
  • the target vector is determined.
  • the determining the target vector based on the first vector and the second vector includes:
  • the performing deformation processing on the region to be adjusted of the object based on the orientation of the reference region includes:
  • a deformation process is performed on the area to be adjusted of the object.
  • the determining the orientation of the area to be adjusted based on the orientation of the reference area includes:
  • the direction of the area to be adjusted is opposite to the direction of the reference area;
  • the first type of area includes: a hip area;
  • the orientation of the area to be adjusted is the same as the orientation of the reference area;
  • the second type of area includes: a face area, a shoulder area, and a crotch area.
  • the performing deformation processing on the region to be adjusted of the object based on the orientation of the reference region includes:
  • the region to be adjusted is deformed, wherein the direction of the first reference vector is perpendicular to the shooting direction of the image, and the first reference vector The direction of the reference vector is opposite to the direction of the second reference vector.
  • the area to be adjusted includes a first sub-area and a second sub-area; the first sub-area and the second sub-area are symmetrical with respect to the center line of the area to be adjusted;
  • the performing deformation processing on the region to be adjusted according to the target vector, the first reference vector, and the second reference vector includes:
  • the first included angle between the vectors and the second included angle between the target vector and the second reference vector adjust the area of the first sub-region and the area of the second sub-region.
  • the direction of the target vector is different from the direction of the first reference vector
  • the direction of the target vector is different from the direction of the second reference vector
  • the first included angle between the target vector and the first reference vector and the second included angle between the target vector and the second reference vector are adjusted to adjust the area of the first sub-region and the The area of the second sub-region includes:
  • the area of the first sub-region is adjusted according to the first area adjustment amount, and the area of the second sub-region is adjusted according to the second area adjustment amount.
  • the first area adjustment amount of the first sub-region and the second area adjustment amount of the second sub-region are determined according to the first included angle and the second included angle Quantity, including:
  • the first area adjustment amount is compared with the second The ratio of the area adjustment amount is taken as the first ratio, and the ratio of the first included angle to the second included angle is taken as the second ratio;
  • the first area adjustment amount and the second area adjustment amount are determined so that the first ratio and the second ratio are positively correlated.
  • the method further includes:
  • the first area adjustment amount is equal to the second area Adjustment amount.
  • an image processing device including:
  • An obtaining unit for obtaining key points of the reference area of the object in the image An obtaining unit for obtaining key points of the reference area of the object in the image
  • a determining unit configured to determine the orientation of the reference area according to the key points of the reference area
  • the processing unit is configured to perform deformation processing on the area to be adjusted of the object based on the orientation of the reference area; the area to be adjusted is the same or different from the reference area.
  • the determining unit is configured to:
  • the at least three key points include a first key point, a second key point, and a third key point; the first key point and the third key point are related to the second key point Point symmetry
  • the determining unit is used to:
  • the target vector is determined.
  • the determining unit is configured to:
  • the processing unit is configured to:
  • a deformation process is performed on the area to be adjusted of the object.
  • the processing unit is configured to:
  • the direction of the area to be adjusted is opposite to the direction of the reference area;
  • the first type of area includes: a hip area;
  • the orientation of the area to be adjusted is the same as the orientation of the reference area;
  • the second type of area includes: a face area, a shoulder area, and a crotch area.
  • the processing unit is configured to:
  • the region to be adjusted is deformed, wherein the direction of the first reference vector is perpendicular to the shooting direction of the image, and the first reference vector The direction of the reference vector is opposite to the direction of the second reference vector.
  • the area to be adjusted includes a first sub-area and a second sub-area; the first sub-area and the second sub-area are symmetrical with respect to the center line of the area to be adjusted;
  • the processing unit is used to:
  • the first included angle between the vectors and the second included angle between the target vector and the second reference vector adjust the area of the first sub-region and the area of the second sub-region.
  • the processing unit is configured to:
  • the area of the first sub-region is adjusted according to the first area adjustment amount, and the area of the second sub-region is adjusted according to the second area adjustment amount.
  • the processing unit is configured to:
  • the first area adjustment amount is compared with the second The ratio of the area adjustment amount is taken as the first ratio, and the ratio of the first included angle to the second included angle is taken as the second ratio;
  • the first area adjustment amount and the second area adjustment amount are determined so that the first ratio and the second ratio are positively correlated.
  • processing unit is further configured to:
  • the first area adjustment amount is equal to the second area adjustment amount.
  • an image processing device including:
  • the processor is connected to the memory and configured to implement the image processing method provided by any of the foregoing technical solutions by executing computer executable instructions stored on the memory.
  • a computer storage medium stores computer executable instructions; the computer executable instructions can implement the image processing method provided by any of the foregoing technical solutions.
  • the orientation of the reference region is acquired, and the region to be adjusted is deformed according to the orientation of the reference region, thus reducing the ignorance of the orientation of the reference region
  • the visual weird deformation effect produced by the deformation directly reduces the appearance of weird deformation and improves the image effect after deformation.
  • FIG. 1 is a schematic diagram of a pixel coordinate system provided by an embodiment of the disclosure
  • FIG. 2 is a schematic flowchart of an image processing method provided by an embodiment of the disclosure
  • FIG. 3 is a schematic diagram of a key point provided by an embodiment of the disclosure.
  • FIG. 4 is a schematic flowchart of another image processing method provided by an embodiment of the disclosure.
  • FIG. 5 is a schematic diagram of determining an orientation based on a vector formed by key points according to an embodiment of the disclosure
  • FIG. 6 is a schematic diagram of a centerline provided by an embodiment of the disclosure.
  • FIG. 7 is a schematic diagram of the effect of a deformation processing provided by an embodiment of the disclosure.
  • FIG. 8 is a schematic diagram of the effect of another deformation processing provided by an embodiment of the disclosure.
  • FIG. 9 is a schematic diagram of another deformation processing effect provided by an embodiment of the disclosure.
  • FIG. 10 is a schematic structural diagram of an image processing device provided by an embodiment of the disclosure.
  • FIG. 11 is a schematic structural diagram of an image device provided by an embodiment of the disclosure.
  • the pixel coordinate system in the embodiment of the present disclosure.
  • the lower left corner of the human body image A is the origin o of the pixel coordinate system
  • the direction parallel to the row of the human body image A is the x-axis direction
  • the direction parallel to the column of the human body image A is the y-axis direction.
  • the abscissa is used to indicate the number of columns of the pixels in the human body image A in the human body image A
  • the ordinate is used to indicate the number of rows of the pixels in the human body image A in the face image A
  • the abscissa and The unit of the ordinate can be pixel.
  • the coordinates of pixel a in Figure 1 are (10, 30), that is, the abscissa of pixel a is 30 pixels, the ordinate of pixel a is 20 pixels, and the pixel a is the 30th column in the face image A. Pixel in row 20.
  • this embodiment provides an image processing method, including:
  • Step S210 Obtain key points of the reference area of the object in the image
  • Step S220 Determine the orientation of the reference area according to the key points of the reference area
  • Step S230 Based on the orientation, deform the region to be adjusted of the object.
  • the image processing method provided in this embodiment can be applied to various types of electronic devices capable of processing images, for example, various user equipment such as mobile phones, tablet computers, or wearable devices.
  • Obtaining the key points of the reference area of the image in step S210 may include:
  • a deep learning model such as a neural network is used to detect the key points of the reference area.
  • the key points may be key points of the skeleton of the reference area.
  • connecting these key points can form the skeleton of the reference area.
  • the key point may be a 2D key point.
  • the key point may be a 3D key point.
  • the 3D image may include: an RGB image and a depth image corresponding to the RGB image, or a YUV image and a depth image corresponding to the YUV image.
  • the pixel value of the depth image may be: the distance value between the camera that collects the RGB image or the YUV image and the collection object, and the pixel value that represents the distance may be called the depth value.
  • a depth camera can collect the 3D image; in addition to the usual 2D camera that collects RGB images or YUV images, the depth camera also includes a depth camera that collects depth images such as Time of Flight (TOF).
  • TOF Time of Flight
  • the 2D key point is (x, y); the 3D key point may be (x, y, z).
  • the coordinates of the 2D key points are coordinates in a plane coordinate system; the coordinates of the 3D key points are coordinates in a 3D coordinate system.
  • Figure 3 is a schematic diagram of the human skeleton.
  • Figure 3 shows a schematic diagram of 17 skeleton key points on the human skeleton, which are numbered from 0 to 16, respectively.
  • the skeleton key point numbered 0 is also called No. 0 key point or root node.
  • Key points 11 and 14 respectively correspond to the two shoulder key points of the human skeleton;
  • key points 1 and 4 respectively correspond to the two key points of the crotch.
  • Key point 7 corresponds to the key point of the torso center.
  • Key points 8 and 9 respectively correspond to the two end points of the neck;
  • key point 10 is the head key point.
  • the area covered by the object in the image includes key point areas, where the objects include people and animals, the key point areas all include key points, and the key point areas include: face area, shoulder area, and crotch area.
  • the reference area is the key point area closest to the area to be adjusted.
  • the first center point is determined according to the coordinates of the center point (hereinafter referred to as the first center point) of the area to be adjusted and the center point of the key point area (hereinafter referred to as the second center point).
  • the distance between one center point and the second center point is used as the distance between the area to be adjusted and the key point area.
  • the shortest distance between the area to be adjusted and the key point area is taken as the distance between the area to be adjusted and the key point area.
  • the orientation of the reference area includes at least one of the following:
  • the crotch area may include the waist and abdomen area.
  • Deforming the area to be adjusted may include: performing pixel transformation on the image area including the area to be adjusted to produce a visual deformation effect.
  • the pixel transformation can be performed using the following methods:
  • the deformed grid is used to assist the deformation processing of the region to be adjusted.
  • the grid points in the deformed grid are control points for deformation processing, and the coordinate change of the control point directly determines the conversion of the pixel coordinates of the pixels in the grid where the frame control point is located.
  • the pixel transformation corresponding to a certain control point can be determined based on the deformation interpolation algorithm.
  • the deformation interpolation algorithm may be a spline curve algorithm.
  • the deformed grid may be a crisscrossed grid, and the intersection of the crisscrossed deformation lines is the control point of the deformed grid.
  • the mapping of coordinates When adjusting the control points, coordinate adjustments in at least two directions can be performed, so that at least the zoom in or zoom out of the area to be adjusted can be achieved, and the zoom in the area to be adjusted can obtain a visual zoom effect. By zooming out, you can get the visual zooming effect of the area to be adjusted.
  • the deformation processing of the area to be adjusted is performed according to the orientation of the reference area, in this way, the deformation processing of the area to be adjusted can be accurately performed according to the orientation of the reference area, instead of directly performing deformation without considering the orientation of the reference area. Processing to reduce the weird deformation processing caused by not considering the orientation of the reference area, thereby improving the quality of the image after the deformation processing.
  • the area to be adjusted is deformed according to the orientation of the area to be adjusted. Since the area to be adjusted may not contain key points, it may not be possible to determine the direction of the area to be adjusted based on the area to be adjusted. Since the reference areas all contain key points, the orientation of the area to be adjusted can be determined according to the orientation of the reference area.
  • the direction opposite to the direction of the reference area may be used as the direction of the area to be adjusted.
  • the area to be adjusted is the hip area and the reference area is the crotch area. Therefore, the direction opposite to the direction of the crotch area may be used as the direction of the area to be adjusted.
  • the orientation of the reference area may be used as the orientation of the area to be adjusted.
  • the area to be adjusted is the chest
  • the reference area is the shoulder area
  • the orientation of the shoulder area can be used as the orientation of the area to be adjusted.
  • the orientation of the reference area in the image is determined according to the key points; when the deformation processing of the area to be adjusted is performed, it is no longer the same deformation processing for each part of the area to be adjusted, but The area to be adjusted will be deformed in different directions according to the orientation.
  • the step S220 may include:
  • Step S221 Obtain at least three key points in the reference area
  • Step S222 Determine a target vector based on the at least three key points
  • the above at least three key points are not on the same straight line.
  • the target vector can be determined based on at least three key points, and the direction of the target vector is used as the orientation of the reference area.
  • the area to be adjusted can be divided into at least two types according to the orientation: the first type area and the second type area, where the direction of the first type area is opposite to the direction of the face area, and the direction of the second type area is the same as that of the face.
  • the areas are facing the same.
  • the first type of area includes: hip area.
  • the second type of area includes: face area, shoulder area, and crotch area.
  • determining the orientation of the area to be adjusted according to the orientation of the reference area may include: when the area to be adjusted is the first type of area, the direction opposite to the orientation of the reference area is taken as the area to be adjusted. Adjust the orientation of the area. When the area to be adjusted is the second type of area, the orientation of the reference area is taken as the orientation of the area to be adjusted.
  • the area to be adjusted is the hip area, and the reference area is the crotch area. Since the area to be adjusted is the first type of area, the direction opposite to the direction of the reference area can be used as the direction of the area to be adjusted.
  • the area to be adjusted is the chest area and the reference area is the shoulder area. Since the area to be adjusted is the second type of area, the orientation of the reference area can be used as the direction of the area to be adjusted.
  • the reference area and the area to be adjusted may be the same or different.
  • the area to be adjusted is a shoulder area, and the reference area can also be a shoulder area.
  • the target vector can be determined based on at least three key points in the shoulder area, and the direction of the area to be adjusted can be determined according to the direction of the target vector.
  • the area to be adjusted is the leg area, and the reference area can also be the crotch area.
  • the target vector can be determined based on at least three key points in the crotch area, and the direction of the target vector is taken as the direction of the crotch area.
  • the orientation of the leg area determines the orientation of the leg area.
  • step S220 may further include:
  • Step S223 Determine the orientation of the region to be adjusted according to the orientation of the target vector.
  • the step S230 may include: deforming the region to be adjusted based on the orientation of the target vector.
  • the area to be adjusted when the area to be adjusted is the first type of area, the opposite direction of the target vector is taken as the direction of the area to be adjusted.
  • the area to be adjusted is the second type of area, the direction of the target vector is taken as the direction of the area to be adjusted.
  • the orientation of the area to be adjusted can be determined according to the orientation of the reference area to improve processing efficiency.
  • the deformation processing of the area to be adjusted can also be performed directly according to the orientation of the reference area, without mapping the orientation of the reference area to the orientation of the area to be adjusted.
  • the orientation of the reference area may be used to perform the deformation processing of the area to be adjusted.
  • the aforementioned at least three key points include a first key point, a second key point, and a third key point, wherein the first key point and the third key point are symmetrical with respect to the second key point.
  • the step S220 may include:
  • the target vector is determined.
  • the first key point, the second key point, and the third key point can form at least two vectors, and the at least two vectors are not in the same straight line, thereby forming a plane.
  • the first key point, the second key point, and the third key point may use the key point 1 of the left crotch shown in FIG. 3 as the first key point.
  • the root node 0 or the key point 7 of the torso center shown in FIG. 3 is taken as the second key point, and the key point 4 of the right hip shown in FIG. 3 is taken as the third key point.
  • the target is obtained based on the first vector (the vector from key point 4 to key point 7) and the second vector (the vector from key point 1 to key point 7). vector.
  • the third key point and the first key point are located on both sides of the second key point in the reference area, and the second key point serves as the third key point and the first key point.
  • Symmetry center of point center symmetry For example, taking the human body as an example, the left and right hips are symmetrically distributed with respect to the center point of the waist, and the corresponding key points of the left and right crotch are symmetrically distributed with respect to the center of the waist.
  • the first vector may be: a vector from key point 4 to key point 7
  • the second vector may be: from key point 1 to The vector of key point 7
  • the third vector may be a vector of key point 7 perpendicular to the plane formed by key point 1, key point 4, and key point 7.
  • the first vector may be a vector from keypoint 4 to keypoint 0, and the second vector may be a vector from keypoint 1 to keypoint 0.
  • the first vector may be: a vector from key point 11 to key point 7
  • the second vector may be: a vector from key point 14 to key point 7. vector.
  • the symmetrical part of the living body is taken as an example.
  • the selected first key point and the third key point are symmetrical with the second key point in the reference area. .
  • a plane is constructed based on the first vector and the second vector, and the normal vector of the plane may be the orientation of the reference area.
  • the determining the orientation of the reference area based on the first vector and the second vector includes:
  • the first vector and the second vector are cross-multiplied to obtain the normal vector of the plane formed by the first vector and the second vector as the target vector.
  • the direction determines the direction of the reference area.
  • the step S230 may include:
  • the region to be adjusted is deformed, wherein the direction of the first reference vector is perpendicular to the shooting direction of the image, and the first reference vector The direction of the reference vector is opposite to the direction of the second reference vector.
  • the aforementioned shooting direction is the direction of the direction vector of the optical axis of the imaging device (including camera, camera, and video camera) that takes the aforementioned image.
  • the shape of the area to be adjusted can be adjusted by performing deformation processing on the area to be adjusted.
  • adjusting the shape of the area to be adjusted includes at least one of the following: enlarging the area of the area to be adjusted, reducing the area of the area to be adjusted, and adjusting the contour of the area to be adjusted.
  • the image containing the user may be deformed to enlarge the buttocks area in the image.
  • the image containing the user can be deformed to reduce the buttocks area in the image.
  • the image containing the user may be deformed to enlarge the breast area in the image.
  • the image containing the user may be deformed to reduce the breast area in the image.
  • the area to be adjusted is a symmetric area
  • the area to be adjusted can be divided into a first sub-area and a second sub-area.
  • the areas to be adjusted in the human body are all symmetric about the center line, that is, the areas to be adjusted are symmetric about the center line.
  • the first subregion is the left hip
  • the second subregion is the right hip
  • the first subregion is the right hip
  • the second subregion is the left hip
  • the first sub-area is the right chest of the human body
  • the second sub-area is the left chest of the human body
  • the left hip, right hip, left chest and right chest of the human body are all distinguished by the left and right parts of the human body.
  • the first sub-area and the second sub-area are the same. Perform the same deformation treatment.
  • the first sub-region and the second sub-region are different (including different areas and different contours). Performing the same deformation processing on one sub-region and the second sub-region will cause the deformed region to be adjusted to appear deformed (for example, the proportion of the deformed region to be adjusted is not consistent). Therefore, the deformation processing for the first subregion and the deformation processing for the second subregion can be determined according to the area of the first subregion and the area of the second subregion to reduce the distortion of the region to be adjusted after the deformation processing. The probability.
  • the angle between the target vector and the first reference vector (hereinafter referred to as the first angle)
  • the angle between the target vector and the second reference vector (hereinafter referred to as The second included angle)
  • the size relationship between the area of the first sub-region and the area of the second sub-region (hereinafter referred to as the area relationship)
  • the deformation processing of the first sub-region and the second sub-region can be determined according to the area relationship. Deformation processing of sub-regions.
  • the deformation processing includes adjusting the area of the region to be adjusted, that is, including adjusting the area of the first sub-region and adjusting the area of the second sub-region.
  • the ratio between the area of the first sub-region and the area of the second sub-region hereinafter referred to as the first area ratio
  • the ratio of the area adjustment amount of the first sub-region to the area adjustment amount of the second sub-region is determined ( This will be referred to as the second area ratio hereinafter).
  • the ratio of the first sub-region after the deformation treatment to the second sub-region after the deformation treatment (including the area of the first sub-region after the deformation treatment and the deformation treatment The ratio of the area of the second sub-region afterwards) is more coordinated, thereby reducing the probability that the region to be adjusted after the deformation processing is deformed, and making the effect of the deformation processing more natural.
  • the area adjustment amount of the first sub-region (hereinafter referred to as the first area adjustment amount) and the area adjustment amount of the second sub-region can be determined (This will be referred to as the second area adjustment amount hereinafter).
  • the area of the first sub-region is adjusted according to the first area adjustment amount
  • the area of the second sub-region is adjusted according to the second area adjustment amount, so as to realize the deformation processing of the area to be adjusted.
  • the first included angle When the first included angle is equal to the second included angle, it represents that the target vector is perpendicular to the first reference vector and the target vector is perpendicular to the second reference vector, that is, the area of the first sub-region is equal to the area of the second sub-region. Therefore, the first area adjustment amount can be made equal to the second area adjustment amount. For example, if the direction of the crotch area is the same as the shooting direction, the first included angle is equal to the second included angle. At this time, the first area adjustment amount and the second area adjustment amount are equal. In this way, if the left hip area expands outward, then The right hip area also expands outward; if the left hip area shrinks inward, the right hip area also shrinks inward.
  • the direction of the target vector is the same as the direction of the first reference vector, or the direction of the target vector is the same as the direction of the second reference vector
  • only one of the first sub-region and the second sub-region is included in the representative image.
  • only the first sub-region or the second sub-region needs to be adjusted. For example, taking the buttocks as an example, if only the right hip area is included in the image, only the right hip area (that is, one of the first and second sub-regions) needs to be adjusted; if only the left hip area is included in the image, You only need to adjust the left hip area.
  • the area of the first subregion is characterized Not equal to the area of the second sub-region.
  • the greater the difference between the first included angle and the second included angle the greater the difference between the area of the first sub-region and the area of the second sub-region.
  • the first area adjustment amount and the second area adjustment can be determined according to the difference between the first included angle and the second included angle (hereinafter referred to as the included angle difference) the amount.
  • a subregion with a large area in the first subregion and a second subregion is regarded as a large subregion
  • a subregion with a small area in the first subregion and the second subregion is regarded as a small subregion.
  • the absolute value of the included angle difference is positively correlated with the area of the large sub-region, and the absolute value of the included angle difference is negatively correlated with the area of the small sub-region.
  • the area adjustment amount of the largest sub-region and the area adjustment amount of the small sub-region can be determined according to the absolute value of the angle difference, so that the area adjustment amount of the large sub-region is equal to
  • the absolute value of the included angle difference is positively correlated, and the area adjustment of the small sub-region is negatively correlated with the absolute value of the included angle difference.
  • the area adjustment amount of the large area the absolute value of the angle difference ⁇ d, where d is a positive number
  • the area adjustment amount of the small area 1/the absolute value of the angle difference.
  • the left chest area is the first sub-area
  • the right chest area is the second sub-area.
  • the first area adjustment amount is greater than the second area adjustment amount
  • the first area adjustment amount is less than The second area adjustment.
  • the left hip area is the first sub-area
  • the right hip area is the second sub-area.
  • the first area adjustment amount is greater than the second area adjustment amount
  • the first area adjustment amount is less than The second area adjustment.
  • the area to be adjusted is the chest and buttocks.
  • the area to be adjusted is not limited to the buttocks and chest.
  • the area to be adjusted may also include: shoulder area, leg area or back Area etc.
  • the implementation of adjusting the area of the area to be adjusted includes at least one of the following: adjusting the area of the area to be adjusted while keeping the shape of the contour of the area to be adjusted unchanged; Adjust the outline of the area to adjust the area of the area to be adjusted.
  • the first area adjustment amount is made larger than the second area adjustment amount, and the area of the first sub-region is smaller than all the areas.
  • making the first area adjustment amount smaller than the second area adjustment amount can make the proportion of the area to be adjusted after the deformation process more coordinated.
  • Figures 7 to 9 of the present disclosure are schematic diagrams of the effect of performing hip-enriching deformation.
  • the hip area of the human body faces the right, and the hip area closer to the right of the human body is hidden in the image.
  • the hip area closer to the left of the human body only needs to be processed for the hip area on the left of the human body, as shown in Figure 8.
  • the left buttock area has a higher uplift in the right image of Figure 8.
  • the orientation of the reference area (ie hip area) shown in FIG. 9 is different from the first reference vector, and the orientation of the reference area is different from the direction of the second reference vector.
  • the angle (the angle between the orientation of the reference area and the first reference vector) and the second angle (the angle between the reference area and the second reference vector) determine the area of the left hip area and the area of the right hip area The relationship between the sizes, and then to achieve the treatment of buttocks.
  • the area adjustment amount of the left hip area is smaller than the area adjustment amount of the right hip area.
  • this embodiment provides an image processing device, including:
  • the acquiring unit 11 is used to acquire the key points of the reference area of the object in the image
  • the determining unit 12 is configured to determine the orientation of the reference area according to the key points of the reference area;
  • the processing unit 13 is configured to perform deformation processing on the area to be adjusted of the object based on the orientation of the reference area; the area to be adjusted is the same or different from the reference area.
  • the determining unit 12 is configured to:
  • the at least three key points include a first key point, a second key point, and a third key point; the first key point and the third key point are related to the second key point Point symmetry
  • the determining unit 12 is configured to:
  • the target vector is determined.
  • the determining unit 12 is configured to:
  • processing unit 13 is configured to:
  • a deformation process is performed on the area to be adjusted of the object.
  • processing unit 13 is configured to:
  • the direction of the area to be adjusted is opposite to the direction of the reference area;
  • the first type of area includes: a hip area;
  • the orientation of the area to be adjusted is the same as the orientation of the reference area;
  • the second type of area includes: a face area, a shoulder area, and a crotch area.
  • processing unit 13 is configured to:
  • the region to be adjusted is deformed, wherein the direction of the first reference vector is perpendicular to the shooting direction of the image, and the first reference vector The direction of the reference vector is opposite to the direction of the second reference vector.
  • the area to be adjusted includes a first sub-area and a second sub-area; the first sub-area and the second sub-area are symmetrical with respect to the center line of the area to be adjusted;
  • the processing unit 13 is configured to:
  • the first included angle between the vectors and the second included angle between the target vector and the second reference vector adjust the area of the first sub-region and the area of the second sub-region.
  • processing unit 13 is configured to:
  • the area of the first sub-region is adjusted according to the first area adjustment amount, and the area of the second sub-region is adjusted according to the second area adjustment amount.
  • processing unit 13 is configured to:
  • the first area adjustment amount is compared with the second The ratio of the area adjustment amount is taken as the first ratio, and the ratio of the first included angle to the second included angle is taken as the second ratio;
  • the first area adjustment amount and the second area adjustment amount are determined so that the first ratio and the second ratio are positively correlated.
  • processing unit 13 is further configured to:
  • the first area adjustment amount is equal to the second area adjustment amount.
  • this embodiment provides an image device, including:
  • the processor is connected to the memory, and is configured to implement the image processing method provided by one or more of the foregoing embodiments by executing computer-executable instructions located on the memory, for example, the image shown in FIG. 2 and FIG. 4 One or more of the processing methods.
  • the memory can be various types of memory, such as random access memory, read-only memory, flash memory, etc.
  • the memory can be used for information storage, for example, storing computer executable instructions and the like.
  • the computer-executable instructions may be various program instructions, for example, target program instructions and/or source program instructions.
  • the processor may be various types of processors, for example, a central processing unit, a microprocessor, a digital signal processor, a programmable array, a digital signal processor, an application specific integrated circuit, or an image processor.
  • the processor may be connected to the memory through a bus.
  • the bus may be an integrated circuit bus or the like.
  • the image device may further include: a communication interface
  • the communication interface may include: a network interface, for example, a local area network interface, a transceiver antenna, and the like.
  • the communication interface is also connected to the processor and can be used for information transmission and reception.
  • the electronic device further includes a human-computer interaction interface.
  • the human-computer interaction interface may include various input and output devices, such as a keyboard, a touch screen, and the like.
  • This embodiment provides a computer storage medium that stores computer-executable instructions; after the computer-executable instructions are executed, the image processing method provided by one or more of the foregoing embodiments can be implemented, for example, 2 and one or more of the image processing methods shown in Figure 4.
  • the computer storage medium may include various recording media with recording functions, for example, various storage media such as CD, floppy disk, hard disk, magnetic tape, optical disk, U disk, or mobile hard disk.
  • the optional computer storage medium may be a non-transitory storage medium, which can be read by the processor, so that the computer executable instructions stored on the computer storage mechanism can be obtained and executed by the processor to realize the foregoing.
  • the information processing method provided by any one of the technical solutions, for example, executes an information processing method applied to a terminal device or an information processing method applied to an application server.
  • This embodiment also provides a computer program product that includes computer-executable instructions; after the computer-executable instructions are executed, the image processing method provided by one or more of the foregoing embodiments can be implemented, for example, 2 and one or more of the image processing methods shown in Figure 4.
  • the computer program includes a computer program tangibly contained on a computer storage medium.
  • the computer program includes program code for executing the method shown in the flowchart.
  • the program code may include instructions corresponding to the steps of the method provided in the embodiments of the present application.
  • the disclosed device and method may be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation, such as: multiple units or components can be combined, or It can be integrated into another system, or some features can be ignored or not implemented.
  • the coupling, or direct coupling, or communication connection between the components shown or discussed may be indirect coupling or communication connection through some interfaces, devices or units, and may be electrical, mechanical or other forms of.
  • the units described above as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units; Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • the functional units in the embodiments of the present disclosure can be all integrated into one processing module, or each unit can be individually used as a unit, or two or more units can be integrated into one unit;
  • the unit can be implemented in the form of hardware, or in the form of hardware plus software functional units.
  • a person of ordinary skill in the art can understand that all or part of the steps in the above method embodiments can be implemented by a program instructing relevant hardware.
  • the foregoing program can be stored in a computer readable storage medium. When the program is executed, it is executed. Including the steps of the foregoing method embodiment; and the foregoing storage medium includes: removable storage devices, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks, etc.
  • ROM read-only memory
  • RAM Random Access Memory
  • magnetic disks or optical disks etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Medical Informatics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Databases & Information Systems (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Quality & Reliability (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Processing Or Creating Images (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

一种图像处理方法、装置、设备及存储介质。所述图像处理方法包括：获取图像中的对象的参考区域的关键点（S210）；根据所述参考区域的关键点，确定所述参考区域的朝向（S220）；基于所述参考区域的朝向，对所述对象的待调整区域进行形变处理（S230）；所述待调整区域与所述参考区域相同或不同。

Description

图像处理方法及装置、图像设备及存储介质
本申请要求于2019年3月14日提交中国专利局、申请号为201910191918.1的中国申请的优先权，以及于2019年3月14日提交中国专利局、申请号为201910193649.2的中国申请的优先权。
技术领域
本公开涉及信息技术领域，尤其涉及一种图像处理方法、装置、设备及存储介质。
背景技术
在对图像中的对象所覆盖的区域进行变形时，一般是基于用户手动操作进行变形，例如，photoshop等图像处理软件，这很大程度上取决于用户的操作技能；对于一般用户而言操作难度大。在相关技术中会出现了另一种图像处理软件，用户进行一键式操作，则图像处理软件就对该图像进行整体处理，但是这种由电子设备的执行操作，由于局限设备处理的机械性(即缺少智能性)，处理后的图像效果有些怪异，图像的处理效果并不如预期。
发明内容
有鉴于此，本公开实施例期望提供一种图像处理方法、装置、设备及存储介质。
第一方面，提供了一种图像处理方法，包括：
获取图像中的对象的参考区域的关键点；
根据所述参考区域的关键点，确定所述参考区域的朝向；
基于所述参考区域的朝向，对所述对象的待调整区域进行形变处理；所述待调整区域与所述参考区域相同或不同。
结合本申请任一实施方式，所述根据所述参考区域的关键点，确定所述参考区域的朝向，包括：
获取所述参考区域中的至少三个关键点；所述至少三个关键点不在同一直线上；
基于所述至少三个关键点，确定目标向量；
将所述目标向量的方向作为所述参考区域的朝向。
结合本申请任一实施方式，所述至少三个关键点包括第一关键点、第二关键点和、第三关键点；所述第一关键点与所述第三关键点关于所述第二关键点对称；
所述基于所述至少三个关键点，确定目标向量，包括：
基于所述第一关键点和所述第二关键点构建第一向量；
基于所述第三关键点和所述第二关键点构建第二向量；
基于所述第一向量和所述第二向量，确定所述目标向量。
结合本申请任一实施方式，所述基于所述第一向量和所述第二向量，确定所述目标向量，包括：
将所述第一向量与所述第二向量叉乘，得到所述目标向量。
结合本申请任一实施方式，所述基于所述参考区域的朝向，对所述对象的待调整区域进行形变处理，包括：
基于所述参考区域的朝向，确定待调整区域的朝向；
基于所述待调整区域的朝向，对所述对象的待调整区域进行形变处理。
结合本申请任一实施方式，所述基于所述参考区域的朝向，确定待调整区域的朝向，包括：
在所述待调整区域属于第一类区域的情况下，所述待调整区域的朝向与所述参考区域的朝向相反；所述第一类区域包括：臀部区域；
在所述待调整区域属于第二类区域的情况下，所述待调整区域的朝向与所述参考区域的朝向相同；所述第二类区域包括：面部区域、肩部区域、胯部区域。
结合本申请任一实施方式，所述基于所述参考区域的朝向，对所述对象的待调整区域进行形变处理，包括：
根据所述目标向量、第一参考向量和第二参考向量，对所述待调整区域进行形变处理，其中，所述第一参考向量的方向与所述图像的拍摄方向垂直，且所述第一参考向量的方向与所述第二参考向量的方向相反。
结合本申请任一实施方式，所述待调整区域包括第一子区域和第二子区域；所述第一子区域和所述第二子区域关于所述待调整区域的中心线对称；
所述根据所述目标向量、第一参考向量和第二参考向量，对所述待调整区域进行形变处理，包括：
在所述目标向量的方向与所述第一参考向量的方向不同，且所述目标向量的方向与所述第二参考向量的方向不同的情况下，根据所述目标向量与所述第一参考向量之间的第一夹角，以及，所述目标向量与所述第二参考向量之间的第二夹角，调整所述第一子区域的面积和所述第二子区域的面积。
结合本申请任一实施方式，所述在所述目标向量的方向与所述第一参考向量的方向不同，且所述目标向量的方向与所述第二参考向量的方向不同的情况下，根据所述目标向量与所述第一参考向量之间的第一夹角以及所述目标向量与所述第二参考向量之间的第二夹角，调整所述第一子区域的面积和所述第二子区域的面积，包括：
根据所述第一夹角和所述第二夹角，确定所述第一子区域的第一面积调整量和所述第二子区域的第二面积调整量；
根据所述第一面积调整量调整所述第一子区域的面积，根据所述第二面积调整量调整所述第二子区域的面积。
结合本申请任一实施方式，所述根据所述第一夹角和所述第二夹角，确定所述第一子区域的第一面积调整量和所述第二子区域的第二面积调整量，包括：
在所述第一夹角大于所述第二夹角，且所述第一子区域的面积大于所述第二子区域的面积的情况下，将所述第一面积调整量与所述第二面积调整量的比值作为第一比值，将所述第一夹角与所述第二夹角的比值作为第二比值；
依据所述第二比值，确定所述第一面积调整量和所述第二面积调整量，使所述第一比值与所述第二比值呈正相关。
结合本申请任一实施方式，所述方法还包括：
在所述第一夹角等于所述第二夹角，且所述第一子区域的面积等于所述第二子区域的面积的情况下，所述第一面积调整量等于所述第二面积调整量。
第二方面，提供了一种图像处理装置，包括：
获取单元，用于获取图像中的对象的参考区域的关键点；
确定单元，用于根据所述参考区域的关键点，确定所述参考区域的朝向；
处理单元，用于基于所述参考区域的朝向，对所述对象的待调整区域进行形变处理；所述待调整区域与所述参考区域相同或不同。
结合本申请任一实施方式，所述确定单元，用于：
获取所述参考区域中的至少三个关键点；所述至少三个关键点不在同一直线上；
基于所述至少三个关键点，确定目标向量；
将所述目标向量的方向作为所述参考区域的朝向。
结合本申请任一实施方式，所述至少三个关键点包括第一关键点、第二关键点、第三关键点；所述第一关键点与所述第三关键点关于所述第二关键点对称；
所述确定单元，用于：
基于所述第一关键点和所述第二关键点构建第一向量；
基于所述第三关键点和所述第二关键点构建第二向量；
基于所述第一向量和所述第二向量，确定所述目标向量。
结合本申请任一实施方式，所述确定单元，用于：
将所述第一向量与所述第二向量叉乘，得到所述目标向量。
结合本申请任一实施方式，所述处理单元，用于：
基于所述参考区域的朝向，确定待调整区域的朝向；
基于所述待调整区域的朝向，对所述对象的待调整区域进行形变处理。
结合本申请任一实施方式，所述处理单元，用于：
在所述待调整区域属于第一类区域的情况下，所述待调整区域的朝向与所述参考区域的朝向相反；所述第一类区域包括：臀部区域；
在所述待调整区域属于第二类区域的情况下，所述待调整区域的朝向与所述参考区域的朝向相同；所述第二类区域包括：面部区域、肩部区域、胯部区域。
结合本申请任一实施方式，所述处理单元，用于：
根据所述目标向量、第一参考向量和第二参考向量，对所述待调整区域进行形变处理，其中，所述第一参考向量的方向与所述图像的拍摄方向垂直，且所述第一参考向量的方向与所述第二参考向量的方向相反。
结合本申请任一实施方式，所述待调整区域包括第一子区域和第二子区域；所述第一子区域和所述第二子区域关于所述待调整区域的中心线对称；
所述处理单元，用于：
在所述目标向量的方向与所述第一参考向量的方向不同，且所述目标向量的方向与所述第二参考向量的方向不同的情况下，根据所述目标向量与所述第一参考向量之间的第一夹角，以及，所述目标向量与所述第二参考向量之间的第二夹角，调整所述第一子区域的面积和所述第二子区域的面积。
结合本申请任一实施方式，所述处理单元，用于：
根据所述第一夹角和所述第二夹角，确定所述第一子区域的第一面积调整量和所述第二子区域的第二面积调整量；
根据所述第一面积调整量调整所述第一子区域的面积，根据所述第二面积调整量调整所述第二子区域的面积。
结合本申请任一实施方式，所述处理单元，用于：
在所述第一夹角大于所述第二夹角，且所述第一子区域的面积大于所述第二子区域的面积的情况下，将所述第一面积调整量与所述第二面积调整量的比值作为第一比值，将所述第一夹角与所述第二夹角的比值作为第二比值；
依据所述第二比值，确定所述第一面积调整量和所述第二面积调整量，使所述第一比值与所述第二比值呈正相关。
结合本申请任一实施方式，所述处理单元，还用于：
在所述第一夹角等于所述第二夹角的情况下，所述第一面积调整量等于所述第二面积调整量。
第三方面，提供了一种图像处理设备，包括：
存储器；
处理器，与所述存储器连接，用于通过执行存储在所述存储器上的计算机可执行指令能够实现前述任意技术方案提供的图像处理方法。
第四方面，提供了一种计算机存储介质，所述计算机存储介质存储有计算机可执行指令；所述计算机可执行指令能够实现前述任意技术方案提供的图像处理方法。
本公开实施例提供的技术方案，在进行待调整区域的变形之前，会获取参考区域的朝向，根据参考区域的朝向来对所述待调整区域进行变形，如此，减少忽略所述参考区域的朝向直接进行形变产生的视觉上的怪异形变效果，从而减少怪异形变的出现，提升了形变后的图像效果。
附图说明
为了更清楚地说明本申请实施例或背景技术中的技术方案，下面将对本申请实施例或背景技术中所需要使用的附图进行说明。
此处的附图被并入说明书中并构成本说明书的一部分，这些附图示出了符合本公开的实施例，并与说明书一起用于说明本公开的技术方案。
图1为本公开实施例提供的一种像素坐标系的示意图；
图2为本公开实施例提供的一种图像处理方法的流程示意图；
图3为本公开实施例提供的一种关键点的示意图；
图4为本公开实施例提供的另一种图像处理方法的流程示意图；
图5为本公开实施例提供的基于关键点形成的向量确定朝向的示意图；
图6为本公开实施例提供的一种中心线的示意图；
图7为本公开实施例提供的一种形变处理的效果示意图；
图8为本公开实施例提供的另一种形变处理的效果示意图；
图9为本公开实施例提供的又一种形变处理效果示意图；
图10为本公开实施例提供的图像处理装置的结构示意图；
图11为本公开实施例提供的图像设备的结构示意图。
具体实施方式
以下结合说明书附图及具体实施例对本公开的技术方案做进一步的详细阐述。
在进行接下来的阐述之前，首先对本公开实施例中的像素坐标系进行定义。如图1所示，以人体图像A的左下角为像素坐标系的原点o、平行于人体图像A的行的方向为x轴的方向、平行于人体图像A的列的方向为y轴的方向，构建像素坐标系xoy。在像素坐标系下，横坐标用于表示人体图像A中的像素在人体图像A中的列数，纵坐标用于表示人体图像A中的像素在人脸图像A中的行数，横坐标和纵坐标的单位均可以是像素。例如，假设图1中的像素a的坐标为(10，30)，即像素a的横坐标为30个像素，像素a的纵坐标为20个像素，像素a为人脸图像A中的第30列第20行的像素。
如图2所示，本实施例提供一种图像处理方法，包括：
步骤S210：获取图像中的对象的参考区域的关键点；
步骤S220：根据所述参考区域的关键点，确定所述参考区域的朝向；
步骤S230：基于所述朝向，对所述对象的待调整区域进行形变处理。
本实施例提供的图像处理方法，可以应用于各类型能够对图像进行处理的电子设备，例如，手机、平板电脑或者可穿戴式设备等各种用户设备。
在步骤S210中获取图像的参考区域的关键点，可包括：
利用神经网络等深度学习模型检测所述参考区域的关键点。例如，所述参考区域为生命体为例，所述关键点可为所述参考区域的骨架的关键点。
若检测到所述关键点之后，连接这些关键点就可以形成所述参考区域的骨架。
若所述图像为2D图像，则所述关键点可为2D关键点。
若所述图像为3D图像，则所述关键点可为3D关键点。
所述3D图像可包括：RGB图像和与RGB图像对应的深度图像构成，或者YUV图像和与所述YUV图像对应的深度图像构成。
所述深度图像的像素值可为：采集所述RGB图像或YUV图像的摄像头与采集对象之间的距离值，这种对距离进行表示的像素值可以称之为深度值。例如，基于深度摄像头可以采集所述3D图像；所述深度摄像头除了平常的采集RGB图像或YUV图像的2D摄像头，还包括如飞行时间(Time of Flight，TOF)等采集深度图像的深度摄像头。
所述2D关键点为(x，y)；所述3D关键点可为(x，y，z)。所述2D关键点的坐标为平面坐标系的坐标；片所述3D关键点的坐标为3D坐标系中的坐标。
图3为人体的骨架示意图。在图3中显示人体骨架上17个骨架关键点的示意图，分别编号0至16，其中，编号为0的骨架关键点，又称为0号关键点或根节点。关键点11及14，分别对应了人体骨架的两个肩部关键点；关键点1及4分别对应了胯部两个关键点。关键点7是对应了躯干中心关键点。关键点8和9分别对应了颈部的两个端点；关键点10为头部关键点。
本实施例中，图像中的对象所覆盖的区域包括关键点区域，其中，对象包括人、动物，关键点区域均包含关键点，关键点区域包括：面部区域、肩部区域、胯部区域。参考区域为距离待调整区域最近的关键点区域。
在一种可能实现的方式中，依据待调整区域的中心点(下文将称为第一中心点)的坐标与关键点区域的中心点(下文将称为第二中心点)的坐标，确定第一中心点与第二中心点之间的距离，作为待调整区域与关键点区域之间的距离。
在另一种可能实现的方式中，将待调整区域与关键点区域之间的最短距离作为待调整区域与关键点区域之间的距离。
以人体为例，所述参考区域的朝向包括以下至少之一：
面部区域的朝向；
肩部区域的朝向；
胯部区域的朝向等。
可选的，胯部区域可包括的腰腹区域。
对待调整区域进行形变可包括：对包含所述待调整区域的图像区域进行像素变换从而产生视觉变形效果。
在一些实施例中，所述像素变换可采用以下方法执行：
利用变形网格来辅助所述待调整区域的形变处理。所述变形网格中的网格点为形变处理的控制点，该控制点的坐标变化直接决定了该框控制点所在网格内像素的像素坐标的转换。
在本实施例中，对应于某一个控制点的像素变换是可以基于变形插值算法确定的。该变形插值算法可为样条曲线算法。
所述变形网格可为纵横交错的网格，纵横交错的变形线的交叉点为变形网格的控制点，通过控制点的坐标映射，就可以控制该控制点所在网格所包含像素的像素坐标的映射。在调整所述控制点时，可以进行至少两个方向上的坐标调整，如此至少可以实现对所述待调整区域的放大或缩小，对待调整区域的放大可获得视觉上的放大效果，对待调整区域的缩小即可获得待调整区域视觉上的缩小效果。在本实施例中，由于会根据参考区域的朝向进行待调整区域的形变处理，如此，可以根据参考区域的朝向精确的进行待调整区域的形变处理，相对于不考虑参考区域的朝向直接进行形变处理，减少因为未考虑参考区域的朝向导致的形变处理怪异，进而提升形变处理后的图像的质量。
在本实施例中，为提升形变处理后的图像的质量，依据待调整区域的朝向对待调整区域进行形变处理。由于待调整区域可能不包含关键点，因此，可能无法依据待调整区域确定待调整区域的朝向。由于参考区域均包含关键点，可依据参考区域的朝向确定待调整区域的朝向。
在一种可能实现的方式中，可将参考区域的朝向的反方向作为待调整区域的朝向。例如，所述待调整区域为臀部区域，参考区域为胯部区域，因此可将胯部区域的朝向的反方向作为待调整区域的朝向。
在另一种可能实现的方式中，可将参考区域的朝向作为待调整区域的朝向。例如，所述待调整区域为胸部，参考区域为肩部区域，因此可将肩部区域的朝向作为待调整区域的朝向。
总之，在本实施例中，会根据关键点确定出所述参考区域在图像中的朝向；在进行待调整区域的形变处理时，不再是对待调整区域的各个部位进行相同形变处理，而是会根据 朝向对待调整区域进行不同方向上的形变处理。
在一些实施例中，如图4所示，所述步骤S220可包括：
步骤S221：获取所述参考区域中的至少三个关键点；
步骤S222：基于所述至少三个关键点，确定目标向量；
上述至少三个关键点不在同一直线上。基于至少三个关键点可确定目标向量，并将目标向量的方向作为参考区域的朝向。
本公开中，依据朝向可将待调整区域划分为至少两类：第一类区域和第二类区域，其中，第一类区域的朝向与面部区域的朝向相反，第二类区域的朝向与面部区域的朝向相同。可选的，第一类区域包括：臀部区域。第二类区域包括：面部区域、肩部区域、胯部区域。
由于参考区域的朝向均与面部区域的朝向相同，依据参考区域的朝向确定待调整区域的朝向可包括：在待调整区域为第一类区域的情况下，将参考区域的朝向的反方向作为待调整区域的朝向。在待调整区域为第二类区域的情况下，将参考区域的朝向作为待调整区域的朝向。
例如，待调整区域为臀部区域，参考区域为胯部区域。由于待调整区域为第一类区域，可将参考区域的朝向的反方向作为待调整区域的方向。又例如，待调整区域为胸部区域，参考区域为肩部区域，由于待调整区域为第二类区域，可将参考区域的朝向作为待调整区域的方向。
需要理解的是，参考区域和待调整区域可以相同，也可以不同。例如，待调整区域为肩部区域，参考区域也可为肩部区域，基于肩部区域内的至少三个关键点可确定目标向量，进而可依据目标向量的朝向确定待调整区域的朝向。又例如，待调整区域为腿部区域，参考区域也可为胯部区域，基于胯部区域内的至少三个关键点可确定目标向量，将目标向量的方向作为胯部区域的朝向，依据胯部区域的朝向确定腿部区域的朝向。
在另一些实施例中，所述步骤S220还可包括：
步骤S223：根据所述目标向量的朝向确定所述待调整区域的朝向。此时，所述步骤S230可包括：基于目标向量的朝向，对所述待调整区域进行形变处理。
在本实施例中，在待调整区域为第一类区域的情况下，将目标向量的反方向作为待调整区域的朝向。在待调整区域为第二类区域的情况下，将目标向量的方向作为待调整区域的朝向。
在本实施例中，在确定出待调整区域的朝向有困难的时候，可以根据参考区域的朝向确定待调整区域的朝向，以提高处理效率。
当然，在一些实施例中，也可以直接根据参考区域的朝向进行待调整区域的形变处理，而不用将参考区域的朝向映射为待调整区域的朝向。在本实施例中，为了减少处理步骤，可以利用参考区域的朝向来进行待调整区域的形变处理。
在一些实施例中，上述至少三个关键点包括第一关键点、第二关键点、第三关键点，其中，第一关键点与第三关键点关于第二关键点对称。所述步骤S220可包括：
基于所述参考区域中的第一关键点和第二关键点构建第一向量；
基于所述参考区域中的第三关键点和第二关键点构建第二向量；
基于所述第一向量和所述第二向量，确定所述目标向量。
所述第一关键点、第二关键点及第三关键点能够构成至少两个向量，这至少两个向量不在同一条直线，从而能够形成一个平面。
例如，以参考区域为人体的胯部区域为例，则所述第一关键点、第二关键点及第三关键点可将图3所示的左胯的关键点1作为第一关键点，将图3所示的根节点0或躯干中心的关键点7作为第二关键点，将图3所示的右胯的关键点4作为第三关键点。如图5所示，在确定胯部的朝向时，基于所述第一向量(关键点4指向关键点7的向量)和第二向量(关键点1指向关键点7的向量)得到所述目标向量。
如图5所示，在参考区域中所述第三关键点和第一关键点位于所述第二关键点的两侧， 且所述第二关键点作为所述第三关键点和第一关键点中心对称的对称中心。例如，以人体为例，左胯和右胯相对于腰部中心点是对称分布，则对应的左胯的关键点和右胯的关键点相对于腰部中心点是对称分布的。
在一些实施例中，如图5所示，若参考区域为胯部，则所述第一向量可为：从关键点4到关键点7的向量；第二向量可为：从关键点1到关键点7的向量；第三向量可为关键点7垂直于关键点1、关键点4及关键点7所形成的平面的向量。
在一些实施例中，若参考区域为胯部，则所述第一向量可为：从关键点4到关键点0的向量，第二向量可为：从关键点1到关键点0的向量。
在还有一些实施例中，若所述参考区域为肩部，则所述第一向量可为：关键点11到关键点7的向量；第二向量可为：关键点14到关键点7的向量。
在本实施例中，是以生命体的对称部位为例，则选择参考区域的关键点时，选择的第一关键点和所述第三关键点在参考区域中是以第二关键点为对称。
在另一些实施例中，基于所述第一向量和所述第二向量构建一个平面，该平面的法向量可为参考区域的朝向。
例如，所述基于所述第一向量和所述第二向量，确定所述参考区域的朝向，包括：
将所述第一向量与所述第二向量叉乘，得到目标向量。
在本实施例中，将所述第一向量与所述第二向量叉乘，可得到由所述第一向量和第二向量所构成平面的法向量，作为目标向量，进而可以根据目标向量的方向确定所述参考区域的朝向。
在另一些实施例中，所述步骤S230可包括：
根据所述目标向量、第一参考向量和第二参考向量，对所述待调整区域进行形变处理，其中，所述第一参考向量的方向与所述图像的拍摄方向垂直，且所述第一参考向量的方向与所述第二参考向量的方向相反。
上述拍摄方向为拍摄上述图像的成像设备(包括：相机、摄像头、摄像机)的光轴的方向向量的方向。
在本实施例中，通过对待调整区域进行形变处理，可调整待调整区域的形状。其中，调整待调整区域的形状包括以下至少一种：放大待调整区域的面积、缩小待调整区域的面积、调整待调整区域的轮廓。
例如，对于身体瘦弱用户而言，可能有丰臀等需求，则可对包含该用户的图像进行形变处理，以放大图像中的臀部区域。再例如，对于身体丰满用户而言，可能有瘦臀等需求，则可对包含该用户的图像进行形变处理，以缩小图像中的臀部区域。
例如，对于身体瘦弱用户而言，可能有丰胸等需求，则可对包含该用户的图像进行形变处理，以放大图像中的胸部区域。再例如，对于身体丰满用户而言，可能有瘦胸等需求，则可对包含该用户的图像进行形变处理，以缩小图像中的胸部区域。
由于待调整区域为对称区域，可将待调整区域分为第一子区域和第二子区域。如图6所示，人体中的待调整区域均关于中心线对称，即待调整区域关于中心线对称。
例如，以所述待调整区域为臀部为例，假设所述第一子区域为左臀，则第二子区域为右臀；假设第一子区域为右臀，则第二子区域为左臀。再例如，以所述待调整区域为胸部，则第一子区域为人体的右胸则第二子区域为人体的左胸；若第一子区域为人体的左胸则第二子区域为人体的右胸。值得注意的是：此处的人体的左臀、右臀、左胸及右胸都是以人体左右部分进行区分的。
在待调整区域的朝向与拍摄方向相同，或待调整区域的朝向与拍摄方向相反的情况下，第一子区域和第二子区域相同，此时可将对第一子区域和第二子区域进行相同的形变处理。在待调整区域的朝向与拍摄方向不同，或待调整区域的朝向与拍摄方向的反方向不同的情况下，第一子区域和第二子区域不同(包括面积不同、轮廓不同)，若对第一子区域和第二子区域进行相同的形变处理，将导致形变处理后的待调整区域出现畸形(如：形变处理 后的待调整区域的比例不协调)。因此，可依据第一子区域的面积和第二子区域的面积，确定对第一子区域进行的形变处理和对第二子区域进行的形变处理，以减少形变处理后的待调整区域出现畸形的概率。
在一种可能实现的方式中，依据目标向量与第一参考向量之间的夹角(下文将称为第一夹角)、目标向量与第二参考向量之间的夹角(下文将称为第二夹角)，可确定第一子区域面积和第二子区域的面积之间的大小关系(下文将称为面积关系)，进而可依据面积关系确定第一子区域的形变处理和第二子区域的形变处理。
本公开实施例中，形变处理包括调整待调整区域的面积，即包括调整第一子区域的面积和调整第二子区域的面积。依据第一子区域的面积与第二子区域的面积之间的比值(下文将称为第一面积比值)，确定第一子区域的面积调整量与第二子区域的面积调整量的比值(下文将称为第二面积比值)。通过使第二面积比值与第一面积比值呈正相关，可使形变处理后的第一子区域与形变处理后的第二子区域的比例(包括形变处理后的第一子区域的面积与形变处理后的第二子区域的面积的比值)更协调，进而减小形变处理后的待调整区域出现畸形的概率，使形变处理的效果更自然。
在一种可能实现的方式中，依据第一夹角和第二夹角，可确定第一子区域的面积调整量(下文将称为第一面积调整量)以及第二子区域的面积调整量(下文将称为第二面积调整量)。根据第一面积调整量调整第一子区域的面积，根据第二面积调整量调整所述第二子区域的面积，进而实现对待调整区域的形变处理。
在第一夹角等于第二夹角的情况下，表征目标向量与第一参考向量垂直且目标向量与第二参考向量垂直，即第一子区域的面积等于第二子区域的面积。因此，可使第一面积调整量等于第二面积调整量。例如，若胯部区域的朝向与拍摄方向相同，则第一夹角等于第二夹角，此时第一面积调整量和第二面积调整量相等，如此，若左臀区域向外扩大，则右臀区域也向外扩大；若左臀区域向内缩小，则右臀区域也向内缩小。
在目标向量的方向与第一参考向量的方向相同，或目标向量的方向与第二参考向量的方向相同的情况下，表征图像中只包含第一子区域和第二子区域中的一个，此时，仅需调整第一子区域或第二子区域。例如，以臀部为例，若在图像中只包含右臀区域，则仅需调整右臀区域(即第一子区域和第二子区域中的一个)；若在图像中只包含左臀区域，则仅需调整左臀区域。
在第一夹角不等于第二夹角，且目标向量的方向与第一参考向量的方向不同，且目标向量的方向与第二参考向量的方向不同的情况下，表征第一子区域的面积不等于第二子区域的面积。第一夹角与第二夹角之间的差异越大，第一子区域的面积与第二子区域的面积之间的差异也越大。为使形变处理后的待调整区域的比例更协调，可依据第一夹角与第二夹角之间的差(下文将称为夹角差)，确定第一面积调整量和第二面积调整量。在一种可能实现的方式中，将第一夹角与第二夹角的比值作为第二比值，则第二比值与夹角差呈正相关，因此可依据第二比值，确定第一面积调整量和第二面积调整量。将第一面积调整量与第二面积调整量的比值作为第一比值。依据第二比值，可确定第一面积调整量和第二面积调整量，使第一比值与第二比值呈正相关。例如，第一比值＝第二比值×c，其中，c为正数。在另一种可能实现的方式中，将第一子区域和第二子区域中面积大的子区域作为大子区域，将第一子区域和第二子区域中面积小的子区域作为小子区域，则夹角差的绝对值与大子区域的面积呈正相关，夹角差的绝对值与小子区域的面积呈负相关。因此可依据夹角差的绝对值，确定第大子区域的面积调整量和小子区域的面积调整量(即第一面积调整量和第二面积调整量)，使大子区域的面积调整量与夹角差的绝对值呈正相关，且使小子区域的面积调整量与夹角差的绝对值呈负相关。例如，大区域的面积调整量＝夹角差的绝对值×d，其中，d为正数，小区域的面积调整量＝1/夹角差的绝对值。
以待调整区域为胸部区域为例，假设左胸区域为第一子区域，右胸区域为第二子区域。在左胸区域的面积大于右胸区域的面积的情况下，第一面积调整量大于第二面积调整量， 在左胸区域的面积小于右胸区域的面积的情况下，第一面积调整量小于第二面积调整量。
以待调整区域为臀部区域为例，假设左臀区域为第一子区域，右臀区域为第二子区域。在左臀区域的面积大于右臀区域的面积的情况下，第一面积调整量大于第二面积调整量，在左臀区域的面积小于右臀区域的面积的情况下，第一面积调整量小于第二面积调整量。
以上是对待调整区域是胸部和臀部进行举例说明，在具体实现时，所述待调整区域不局限于臀部和胸部，例如，所述待调整区域还可包括：肩部区域、腿部区域或背部区域等。
需要理解的是，本实施例中，调整待调整区域的面积的实现方式包括以下至少一种：在使待调整区域的轮廓的形状不变的情况下，调整待调整区域的面积；通过调整待调整区域的轮廓调整待调整区域的面积。
总之，在所述第一子区域的面积大于所述第二子区域的面积的情况下，使第一面积调整量大于所述第二面积调整量，在所述第一子区域的面积小于所述第二子区域的面积的情况下，使所述第一面积调整量小于所述第二面积调整量，可使形变处理后的待调整区域的比例更协调。
本公开图7至图9为进行丰臀形变的效果示意图。
在图7中人体的胯部区域朝向左边，由于更加靠近左边的臀部区域在图像中被隐藏，仅需要对人体更靠近右边的臀部区域进行丰臀处理，如图7所示，通过丰臀处理(即上述形变处理)之后，图7右图中明显可见右臀区域隆起更高。
在图8中人体的胯部区域朝向右边，人体更靠近右边的臀部区域在图像中被隐藏起来了，人体的更靠近左边仅需要对人体左边的臀部区域进行丰臀处理，如图8所示，通过丰臀处理之后，图8右图中明显可见左臀区域隆起更高。
图9所示的参考区域(即臀部区域)的朝向与第一参考向量不同，且参考区域的朝向与第二参考向量的方向不同，为了避免丰臀处理导致臀部区域出现畸形，可根据第一夹角(参考区域的朝向与第一参考向量之间的夹角)和第二夹角(参考区域与第二参考向量之间的夹角)，确定左臀区域的面积与右臀区域的面积之间的大小关系，进而实现丰臀处理。在图9中，由于左臀区域的面积小于右臀区域的面积，左臀区域的面积调整量小于右臀区域的面积调整量。
如图10所示，本实施例提供一种图像处理装置，包括：
获取单元11，用于获取图像中的对象的参考区域的关键点；
确定单元12，用于根据所述参考区域的关键点，确定所述参考区域的朝向；
处理单元13，用于基于所述参考区域的朝向，对所述对象的待调整区域进行形变处理；所述待调整区域与所述参考区域相同或不同。
结合本申请任一实施方式，所述确定单元12，用于：
获取所述参考区域中的至少三个关键点；所述至少三个关键点不在同一直线上；
基于所述至少三个关键点，确定目标向量；
将所述目标向量的方向作为所述参考区域的朝向。
结合本申请任一实施方式，所述至少三个关键点包括第一关键点、第二关键点、第三关键点；所述第一关键点与所述第三关键点关于所述第二关键点对称；
所述确定单元12，用于：
基于所述第一关键点和所述第二关键点构建第一向量；
基于所述第三关键点和所述第二关键点构建第二向量；
基于所述第一向量和所述第二向量，确定所述目标向量。
结合本申请任一实施方式，所述确定单元12，用于：
将所述第一向量与所述第二向量叉乘，得到所述目标向量。
结合本申请任一实施方式，所述处理单元13，用于：
基于所述参考区域的朝向，确定待调整区域的朝向；
基于所述待调整区域的朝向，对所述对象的待调整区域进行形变处理。
结合本申请任一实施方式，所述处理单元13，用于：
在所述待调整区域属于第一类区域的情况下，所述待调整区域的朝向与所述参考区域的朝向相反；所述第一类区域包括：臀部区域；
在所述待调整区域属于第二类区域的情况下，所述待调整区域的朝向与所述参考区域的朝向相同；所述第二类区域包括：面部区域、肩部区域、胯部区域。
结合本申请任一实施方式，所述处理单元13，用于：
根据所述目标向量、第一参考向量和第二参考向量，对所述待调整区域进行形变处理，其中，所述第一参考向量的方向与所述图像的拍摄方向垂直，且所述第一参考向量的方向与所述第二参考向量的方向相反。
结合本申请任一实施方式，所述待调整区域包括第一子区域和第二子区域；所述第一子区域和所述第二子区域关于所述待调整区域的中心线对称；
所述处理单元13，用于：
在所述目标向量的方向与所述第一参考向量的方向不同，且所述目标向量的方向与所述第二参考向量的方向不同的情况下，根据所述目标向量与所述第一参考向量之间的第一夹角，以及，所述目标向量与所述第二参考向量之间的第二夹角，调整所述第一子区域的面积和所述第二子区域的面积。
结合本申请任一实施方式，所述处理单元13，用于：
根据所述第一夹角和所述第二夹角，确定所述第一子区域的第一面积调整量和所述第二子区域的第二面积调整量；
根据所述第一面积调整量调整所述第一子区域的面积，根据所述第二面积调整量调整所述第二子区域的面积。
结合本申请任一实施方式，所述处理单元13，用于：
在所述第一夹角大于所述第二夹角，且所述第一子区域的面积大于所述第二子区域的面积的情况下，将所述第一面积调整量与所述第二面积调整量的比值作为第一比值，将所述第一夹角与所述第二夹角的比值作为第二比值；
依据所述第二比值，确定所述第一面积调整量和所述第二面积调整量，使所述第一比值与所述第二比值呈正相关。
结合本申请任一实施方式，所述处理单元13，还用于：
在所述第一夹角等于所述第二夹角的情况下，所述第一面积调整量等于所述第二面积调整量。
如图11所示，本实施例提供了一种图像设备，包括：
存储器；
处理器，与所述存储器连接，用于通过执行位于所述存储器上的计算机可执行指令，能够实现前述一个或多个实施例提供的图像处理方法，例如，图2、及图4所示图像处理方法中的一个或多个。
该存储器可为各种类型的存储器，可为随机存储器、只读存储器、闪存等。所述存储器可用于信息存储，例如，存储计算机可执行指令等。所述计算机可执行指令可为各种程序指令，例如，目标程序指令和/或源程序指令等。
所述处理器可为各种类型的处理器，例如，中央处理器、微处理器、数字信号处理器、可编程阵列、数字信号处理器、专用集成电路或图像处理器等。
所述处理器可以通过总线与所述存储器连接。所述总线可为集成电路总线等。
在一些实施例中，所述图像设备还可包括：通信接口，该通信接口可包括：网络接口、例如，局域网接口、收发天线等。所述通信接口同样与所述处理器连接，能够用于信息收发。
在一些实施例中，所述电子设备还包括人机交互接口，例如，所述人机交互接口可包括各种输入输出设备，例如，键盘、触摸屏等。
本实施例提供一种计算机存储介质，所述计算机存储介质存储有计算机可执行指令；所述计算机可执行指令被执行后，能够实现前述一个或多个实施例提供的图像处理方法，例如，图2及图4所示图像处理方法中的一个或多个。
所述计算机存储介质可为包括具有记录功能的各种记录介质，例如，CD、软盘、硬盘、磁带、光盘、U盘或移动硬盘等各种存储介质。可选的所述计算机存储介质可为非瞬间存储介质，该计算机存储介质可被处理器读取，从而使得存储在计算机存储机制上的计算机可执行指令被处理器获取并执行后，能够实现前述任意一个技术方案提供的信息处理方法，例如，执行应用于终端设备中的信息处理方法或应用服务器中的信息处理方法。
本实施例还提供一种计算机程序产品，所述计算机程序产品包括计算机可执行指令；所述计算机可执行指令被执行后，能够实现前述一个或多个实施例提供的图像处理方法，例如，图2及图4所示图像处理方法中的一个或多个。
所述包括有形地包含在计算机存储介质上的计算机程序，计算机程序包含用于执行流程图所示的方法的程序代码，程序代码可包括对应执行本申请实施例提供的方法步骤对应的指令。
在本申请所提供的几个实施例中，应该理解到，所揭露的设备和方法，可以通过其它的方式实现。以上所描述的设备实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，如：多个单元或组件可以结合，或可以集成到另一个系统，或一些特征可以忽略，或不执行。另外，所显示或讨论的各组成部分相互之间的耦合、或直接耦合、或通信连接可以是通过一些接口，设备或单元的间接耦合或通信连接，可以是电性的、机械的或其它形式的。
上述作为分离部件说明的单元可以是、或也可以不是物理上分开的，作为单元显示的部件可以是、或也可以不是物理单元，即可以位于一个地方，也可以分布到多个网络单元上；可以根据实际的需要选择其中的部分或全部单元来实现本实施例方案的目的。
另外，在本公开各实施例中的各功能单元可以全部集成在一个处理模块中，也可以是各单元分别单独作为一个单元，也可以两个或两个以上单元集成在一个单元中；上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。
本领域普通技术人员可以理解：实现上述方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成，前述的程序可以存储于一计算机可读取存储介质中，该程序在执行时，执行包括上述方法实施例的步骤；而前述的存储介质包括：移动存储设备、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本公开的具体实施方式，但本公开的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本公开揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本公开的保护范围之内。因此，本公开的保护范围应以所述权利要求的保护范围为准。

Claims (24)

  1. 一种图像处理方法，其特征在于，包括：
    获取图像中的对象的参考区域的关键点；
    根据所述参考区域的关键点，确定所述参考区域的朝向；
    基于所述参考区域的朝向，对所述对象的待调整区域进行形变处理；所述待调整区域与所述参考区域相同或不同。
  2. 根据权利要求1所述的方法，其特征在于，所述根据所述参考区域的关键点，确定所述参考区域的朝向，包括：
    获取所述参考区域中的至少三个关键点；所述至少三个关键点不在同一直线上；
    基于所述至少三个关键点，确定目标向量；
    将所述目标向量的方向作为所述参考区域的朝向。
  3. 根据权利要求2所述的方法，其特征在于，所述至少三个关键点包括第一关键点、第二关键点、第三关键点；所述第一关键点与所述第三关键点关于所述第二关键点对称；
    所述基于所述至少三个关键点，确定目标向量，包括：
    基于所述第一关键点和所述第二关键点构建第一向量；
    基于所述第三关键点和所述第二关键点构建第二向量；
    基于所述第一向量和所述第二向量，确定所述目标向量。
  4. 根据权利要求3所述的方法，其特征在于，所述基于所述第一向量和所述第二向量，确定所述目标向量，包括：
    将所述第一向量与所述第二向量叉乘，得到所述目标向量。
  5. 根据权利要求1至4中任意一项所述的方法，其特征在于，所述基于所述参考区域的朝向，对所述对象的待调整区域进行形变处理，包括：
    基于所述参考区域的朝向，确定待调整区域的朝向；
    基于所述待调整区域的朝向，对所述对象的待调整区域进行形变处理。
  6. 根据权利要求5所述的方法，其特征在于，所述基于所述参考区域的朝向，确定待调整区域的朝向，包括：
    在所述待调整区域属于第一类区域的情况下，所述待调整区域的朝向与所述参考区域的朝向相反；所述第一类区域包括：臀部区域；
    在所述待调整区域属于第二类区域的情况下，所述待调整区域的朝向与所述参考区域的朝向相同；所述第二类区域包括：面部区域、肩部区域、胯部区域。
  7. 根据权利要求2至4任一项所述的方法，其特征在于，所述基于所述参考区域的朝向，对所述对象的待调整区域进行形变处理，包括：
    根据所述目标向量、第一参考向量和第二参考向量，对所述待调整区域进行形变处理，其中，所述第一参考向量的方向与所述图像的拍摄方向垂直，且所述第一参考向量的方向与所述第二参考向量的方向相反。
  8. 根据权利要求7所述的方法，其特征在于，所述待调整区域包括第一子区域和第二子区域；所述第一子区域和所述第二子区域关于所述待调整区域的中心线对称；
    所述根据所述目标向量、第一参考向量和第二参考向量，对所述待调整区域进行形变处理，包括：
    在所述目标向量的方向与所述第一参考向量的方向不同，且所述目标向量的方向与所述第二参考向量的方向不同的情况下，根据所述目标向量与所述第一参考向量之间的第一夹角，以及，所述目标向量与所述第二参考向量之间的第二夹角，调整所述第一子区域的面积和所述第二子区域的面积。
  9. 根据权利要求8所述的方法，其特征在于，所述在所述目标向量的方向与所述第一参考向量的方向不同，且所述目标向量的方向与所述第二参考向量的方向不同的情况下，根据所述目标向量与所述第一参考向量之间的第一夹角以及所述目标向量与所述第二参考 向量之间的第二夹角，调整所述第一子区域的面积和所述第二子区域的面积，包括：
    根据所述第一夹角和所述第二夹角，确定所述第一子区域的第一面积调整量和所述第二子区域的第二面积调整量；
    根据所述第一面积调整量调整所述第一子区域的面积，根据所述第二面积调整量调整所述第二子区域的面积。
  10. 根据权利要求9所述的方法，其特征在于，所述根据所述第一夹角和所述第二夹角，确定所述第一子区域的第一面积调整量和所述第二子区域的第二面积调整量，包括：
    在所述第一夹角大于所述第二夹角，且所述第一子区域的面积大于所述第二子区域的面积的情况下，将所述第一面积调整量与所述第二面积调整量的比值作为第一比值，将所述第一夹角与所述第二夹角的比值作为第二比值；
    依据所述第二比值，确定所述第一面积调整量和所述第二面积调整量，使所述第一比值与所述第二比值呈正相关。
  11. 根据权利要求10所述的方法，其特征在于，所述方法还包括：
    在所述第一夹角等于所述第二夹角的情况下，所述第一面积调整量等于所述第二面积调整量。
  12. 一种图像处理装置，其特征在于，包括：
    获取单元，用于获取图像中的对象的参考区域的关键点；
    确定单元，用于根据所述参考区域的关键点，确定所述参考区域的朝向；
    处理单元，用于基于所述参考区域的朝向，对所述对象的待调整区域进行形变处理；所述待调整区域与所述参考区域相同或不同。
  13. 根据权利要求12所述的装置，其特征在于，所述确定单元，用于：
    获取所述参考区域中的至少三个关键点；所述至少三个关键点不在同一直线上；
    基于所述至少三个关键点，确定目标向量；
    将所述目标向量的方向作为所述参考区域的朝向。
  14. 根据权利要求13所述的装置，其特征在于，所述至少三个关键点包括第一关键点、第二关键点、第三关键点；所述第一关键点与所述第三关键点关于所述第二关键点对称；
    所述确定单元，用于：
    基于所述第一关键点和所述第二关键点构建第一向量；
    基于所述第三关键点和所述第二关键点构建第二向量；
    基于所述第一向量和所述第二向量，确定所述目标向量。
  15. 根据权利要求14所述的装置，其特征在于，所述确定单元，用于：
    将所述第一向量与所述第二向量叉乘，得到所述目标向量。
  16. 根据权利要求12至15中任意一项所述的装置，其特征在于，所述处理单元，用于：
    基于所述参考区域的朝向，确定待调整区域的朝向；
    基于所述待调整区域的朝向，对所述对象的待调整区域进行形变处理。
  17. 根据权利要求16所述的装置，其特征在于，所述处理单元，用于：
    在所述待调整区域属于第一类区域的情况下，所述待调整区域的朝向与所述参考区域的朝向相反；所述第一类区域包括：臀部区域；
    在所述待调整区域属于第二类区域的情况下，所述待调整区域的朝向与所述参考区域的朝向相同；所述第二类区域包括：面部区域、肩部区域、胯部区域。
  18. 根据权利要求13至15任一项所述的装置，其特征在于，所述处理单元，用于：
    根据所述目标向量、第一参考向量和第二参考向量，对所述待调整区域进行形变处理，其中，所述第一参考向量的方向与所述图像的拍摄方向垂直，且所述第一参考向量的方向与所述第二参考向量的方向相反。
  19. 根据权利要求18所述的装置，其特征在于，所述待调整区域包括第一子区域和第 二子区域；所述第一子区域和所述第二子区域关于所述待调整区域的中心线对称；
    所述处理单元，用于：
    在所述目标向量的方向与所述第一参考向量的方向不同，且所述目标向量的方向与所述第二参考向量的方向不同的情况下，根据所述目标向量与所述第一参考向量之间的第一夹角，以及，所述目标向量与所述第二参考向量之间的第二夹角，调整所述第一子区域的面积和所述第二子区域的面积。
  20. 根据权利要求18所述的装置，其特征在于，所述处理单元，用于：
    根据所述第一夹角和所述第二夹角，确定所述第一子区域的第一面积调整量和所述第二子区域的第二面积调整量；
    根据所述第一面积调整量调整所述第一子区域的面积，根据所述第二面积调整量调整所述第二子区域的面积。
  21. 根据权利要求20所述的装置，其特征在于，所述处理单元，用于：
    在所述第一夹角大于所述第二夹角，且所述第一子区域的面积大于所述第二子区域的面积的情况下，将所述第一面积调整量与所述第二面积调整量的比值作为第一比值，将所述第一夹角与所述第二夹角的比值作为第二比值；
    依据所述第二比值，确定所述第一面积调整量和所述第二面积调整量，使所述第一比值与所述第二比值呈正相关。
  22. 根据权利要求21所述的装置，其特征在于，所述处理单元，还用于：
    在所述第一夹角等于所述第二夹角的情况下，所述第一面积调整量等于所述第二面积调整量。
  23. 一种图像处理设备，其特征在于，包括：
    存储器；
    处理器，与所述存储器连接，用于通过执行存储在所述存储器上的计算机可执行指令能够实现权利要求1至11任一项提供的图像处理方法。
  24. 一种计算机存储介质，所述计算机存储介质存储有计算机可执行指令；所述计算机可执行指令能够实现权利要求1至11任一项提供的图像处理方法。
PCT/CN2019/130970 2019-01-18 2019-12-31 图像处理方法及装置、图像设备及存储介质 Ceased WO2020181900A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2020558530A JP7109585B2 (ja) 2019-01-18 2019-12-31 画像処理方法及び装置、画像デバイス並びに記憶媒体
SG11202010399VA SG11202010399VA (en) 2019-01-18 2019-12-31 Image processing method and apparatus, image device, and storage medium
US17/073,769 US11538207B2 (en) 2019-01-18 2020-10-19 Image processing method and apparatus, image device, and storage medium

Applications Claiming Priority (5)

Application Number Priority Date Filing Date Title
CN201910049830 2019-01-18
CN201910193649.2A CN111460871B (zh) 2019-01-18 2019-03-14 图像处理方法及装置、存储介质
CN201910191918.1 2019-03-14
CN201910193649.2 2019-03-14
CN201910191918.1A CN111460870A (zh) 2019-01-18 2019-03-14 目标的朝向确定方法及装置、电子设备及存储介质

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/073,769 Continuation US11538207B2 (en) 2019-01-18 2020-10-19 Image processing method and apparatus, image device, and storage medium

Publications (1)

Publication Number Publication Date
WO2020181900A1 true WO2020181900A1 (zh) 2020-09-17

Family

ID=71679913

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/130970 Ceased WO2020181900A1 (zh) 2019-01-18 2019-12-31 图像处理方法及装置、图像设备及存储介质

Country Status (6)

Country Link
US (5) US11538207B2 (zh)
JP (4) JP7109585B2 (zh)
KR (4) KR20210011425A (zh)
CN (7) CN111460871B (zh)
SG (5) SG11202010399VA (zh)
WO (1) WO2020181900A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114333228A (zh) * 2020-09-30 2022-04-12 北京君正集成电路股份有限公司 一种婴儿的智能视频看护方法
CN115564689A (zh) * 2022-10-08 2023-01-03 上海宇勘科技有限公司 一种基于按块处理技术的人工智能图像处理方法及系统

Families Citing this family (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110910393B (zh) * 2018-09-18 2023-03-24 北京市商汤科技开发有限公司 数据处理方法及装置、电子设备及存储介质
CN111476060B (zh) * 2019-01-23 2025-09-05 北京奇虎科技有限公司 人脸清晰度分析方法、装置、计算机设备及存储介质
US11610414B1 (en) * 2019-03-04 2023-03-21 Apple Inc. Temporal and geometric consistency in physical setting understanding
US10902618B2 (en) * 2019-06-14 2021-01-26 Electronic Arts Inc. Universal body movement translation and character rendering system
CA3154216A1 (en) * 2019-10-11 2021-04-15 Beyeonics Surgical Ltd. System and method for improved electronic assisted medical procedures
KR102610840B1 (ko) * 2019-12-19 2023-12-07 한국전자통신연구원 사용자 모션 자동 인식 시스템 및 그 방법
CN111105348A (zh) * 2019-12-25 2020-05-05 北京市商汤科技开发有限公司 图像处理方法及装置、图像处理设备及存储介质
US11972353B2 (en) 2020-01-22 2024-04-30 Electronic Arts Inc. Character controllers using motion variational autoencoders (MVAEs)
US11504625B2 (en) 2020-02-14 2022-11-22 Electronic Arts Inc. Color blindness diagnostic system
US11232621B2 (en) 2020-04-06 2022-01-25 Electronic Arts Inc. Enhanced animation generation based on conditional modeling
US11648480B2 (en) 2020-04-06 2023-05-16 Electronic Arts Inc. Enhanced pose generation based on generative modeling
CN111881838B (zh) * 2020-07-29 2023-09-26 清华大学 具有隐私保护功能的运动障碍评估录像分析方法及设备
US11410366B2 (en) * 2020-09-18 2022-08-09 Unity Technologies Sf Systems and methods for generating a skull surface for computer animation
CN112165630B (zh) * 2020-10-16 2022-11-15 广州虎牙科技有限公司 图像渲染方法、装置、电子设备及存储介质
US12254561B2 (en) * 2021-01-27 2025-03-18 Spreeai Corporation Producing a digital image representation of a body
CN112932468A (zh) * 2021-01-26 2021-06-11 京东方科技集团股份有限公司 一种肌肉运动能力的监测系统以及监测方法
CN115205889A (zh) * 2021-04-13 2022-10-18 影石创新科技股份有限公司 人体朝向识别方法、视频剪辑方法、装置和计算机设备
US11887232B2 (en) 2021-06-10 2024-01-30 Electronic Arts Inc. Enhanced system for generation of facial models and animation
US12236510B2 (en) 2021-06-10 2025-02-25 Electronic Arts Inc. Enhanced system for generation of facial models and animation
US12169889B2 (en) 2021-06-10 2024-12-17 Electronic Arts Inc. Enhanced system for generation of facial models and animation
US20230177881A1 (en) * 2021-07-06 2023-06-08 KinTrans, Inc. Automatic body movement recognition and association system including smoothing, segmentation, similarity, pooling, and dynamic modeling
CN113706699B (zh) 2021-10-27 2022-02-08 腾讯科技（深圳）有限公司 数据处理方法、装置、电子设备及计算机可读存储介质
US20230137337A1 (en) * 2021-10-28 2023-05-04 Texas Instruments Incorporated Enhanced machine learning model for joint detection and multi person pose estimation
EP4429253A4 (en) * 2021-11-04 2025-01-29 Sony Group Corporation Distribution device, distribution method, and program
CN114115544B (zh) * 2021-11-30 2024-01-05 杭州海康威视数字技术股份有限公司 人机交互方法、三维显示设备及存储介质
KR20230090852A (ko) 2021-12-15 2023-06-22 삼성전자주식회사 복수의 카메라를 이용하여 촬영된 손의 3차원 골격 데이터를 획득하는 전자 장치 및 방법
KR102778301B1 (ko) * 2022-02-14 2025-03-11 (주)코어센스 지면반력 데이터를 통한 모션 캡쳐 시스템 보정 방법
CN114782245A (zh) * 2022-05-25 2022-07-22 周佳庆 隐私保护的智能识别方法、装置、存储介质以及电子设备
CN117315201A (zh) * 2022-06-20 2023-12-29 香港教育大学 用于在虚拟世界中动画化化身的系统
JP2024007270A (ja) * 2022-07-05 2024-01-18 大成建設株式会社 操作監視システムおよび操作監視方法
US12387409B2 (en) 2022-10-21 2025-08-12 Electronic Arts Inc. Automated system for generation of facial animation rigs
US12569761B2 (en) 2023-06-01 2026-03-10 Electronic Arts Inc. System for generating visual content within a game application environment
CN116726505A (zh) * 2023-07-13 2023-09-12 深圳十米网络科技有限公司 基于苹果手表的html5体感游戏方法
US12456245B2 (en) 2023-09-29 2025-10-28 Electronic Arts Inc. Enhanced system for generation and optimization of facial models and animation
TWI865289B (zh) * 2024-01-05 2024-12-01 宏碁股份有限公司 手部姿態調整系統、其手部姿態調整方法及其手部角度取得方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180225842A1 (en) * 2016-01-21 2018-08-09 Tencent Technology (Shenzhen) Company Limited Method and apparatus for determining facial pose angle, and computer storage medium
CN108830200A (zh) * 2018-05-31 2018-11-16 北京市商汤科技开发有限公司 一种图像处理方法、装置和计算机存储介质
CN109117726A (zh) * 2018-07-10 2019-01-01 深圳超多维科技有限公司 一种识别认证方法、装置、系统及存储介质
CN109146769A (zh) * 2018-07-24 2019-01-04 北京市商汤科技开发有限公司 图像处理方法及装置、图像处理设备及存储介质
CN109242789A (zh) * 2018-08-21 2019-01-18 成都旷视金智科技有限公司 图像处理方法、图像处理装置及存储介质
CN109376671A (zh) * 2018-10-30 2019-02-22 北京市商汤科技开发有限公司 图像处理方法、电子设备及计算机可读介质

Family Cites Families (101)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4561066A (en) 1983-06-20 1985-12-24 Gti Corporation Cross product calculator with normalized output
JPH0816820A (ja) * 1994-04-25 1996-01-19 Fujitsu Ltd ３次元アニメーション作成装置
US6657628B1 (en) 1999-11-24 2003-12-02 Fuji Xerox Co., Ltd. Method and apparatus for specification, control and modulation of social primitives in animated characters
JP2002024807A (ja) 2000-07-07 2002-01-25 National Institute Of Advanced Industrial & Technology 物体運動追跡手法及び記録媒体
JP4077622B2 (ja) * 2001-11-15 2008-04-16 独立行政法人科学技術振興機構 ３次元人物動画像生成システム
US7215828B2 (en) * 2002-02-13 2007-05-08 Eastman Kodak Company Method and system for determining image orientation
US9177387B2 (en) 2003-02-11 2015-11-03 Sony Computer Entertainment Inc. Method and apparatus for real time motion capture
US20160098095A1 (en) 2004-01-30 2016-04-07 Electronic Scripting Products, Inc. Deriving Input from Six Degrees of Freedom Interfaces
JP2007004732A (ja) * 2005-06-27 2007-01-11 Matsushita Electric Ind Co Ltd 画像生成装置及び画像生成方法
US7869646B2 (en) 2005-12-01 2011-01-11 Electronics And Telecommunications Research Institute Method for estimating three-dimensional position of human joint using sphere projecting technique
US7859540B2 (en) 2005-12-22 2010-12-28 Honda Motor Co., Ltd. Reconstruction, retargetting, tracking, and estimation of motion for articulated systems
JP4148281B2 (ja) * 2006-06-19 2008-09-10 ソニー株式会社 モーションキャプチャ装置及びモーションキャプチャ方法、並びにモーションキャプチャプログラム
JP5076744B2 (ja) * 2007-08-30 2012-11-21 セイコーエプソン株式会社 画像処理装置
JP5229910B2 (ja) 2008-08-08 2013-07-03 株式会社メイクソフトウェア 画像処理装置、画像出力装置、画像処理方法及びコンピュータプログラム
US8253801B2 (en) 2008-12-17 2012-08-28 Sony Computer Entertainment Inc. Correcting angle error in a tracking system
US8588465B2 (en) * 2009-01-30 2013-11-19 Microsoft Corporation Visual target tracking
US8682028B2 (en) 2009-01-30 2014-03-25 Microsoft Corporation Visual target tracking
US8267781B2 (en) * 2009-01-30 2012-09-18 Microsoft Corporation Visual target tracking
CN101930284B (zh) * 2009-06-23 2014-04-09 腾讯科技（深圳）有限公司 一种实现视频和虚拟网络场景交互的方法、装置和系统
KR101616926B1 (ko) * 2009-09-22 2016-05-02 삼성전자주식회사 영상 처리 장치 및 방법
WO2011045768A2 (en) 2009-10-15 2011-04-21 Yeda Research And Development Co. Ltd. Animation of photo-images via fitting of combined models
RU2534892C2 (ru) 2010-04-08 2014-12-10 Самсунг Электроникс Ко., Лтд. Устройство и способ для захвата безмаркерных движений человека
US9177409B2 (en) 2010-04-29 2015-11-03 Naturalmotion Ltd Animating a virtual object within a virtual world
US8437506B2 (en) * 2010-09-07 2013-05-07 Microsoft Corporation System for fast, probabilistic skeletal tracking
TWI534756B (zh) * 2011-04-29 2016-05-21 國立成功大學 運動編碼影像及生成模組、影像處理模組及運動展示模組
AU2011203028B1 (en) * 2011-06-22 2012-03-08 Microsoft Technology Licensing, Llc Fully automatic dynamic articulated model calibration
US9134127B2 (en) * 2011-06-24 2015-09-15 Trimble Navigation Limited Determining tilt angle and tilt direction using image processing
US10319133B1 (en) 2011-11-13 2019-06-11 Pixar Posing animation hierarchies with dynamic posing roots
KR101849373B1 (ko) 2012-01-31 2018-04-17 한국전자통신연구원 인체의 관절구조를 추정하기 위한 장치 및 방법
US9747495B2 (en) 2012-03-06 2017-08-29 Adobe Systems Incorporated Systems and methods for creating and distributing modifiable animated video messages
CN102824176B (zh) 2012-09-24 2014-06-04 南通大学 一种基于Kinect传感器的上肢关节活动度测量方法
US10241639B2 (en) 2013-01-15 2019-03-26 Leap Motion, Inc. Dynamic user interactions for display control and manipulation of display objects
TWI475495B (zh) * 2013-02-04 2015-03-01 Wistron Corp 圖像的識別方法、電子裝置與電腦程式產品
US9766709B2 (en) * 2013-03-15 2017-09-19 Leap Motion, Inc. Dynamic user interactions for display control
CN104103090A (zh) * 2013-04-03 2014-10-15 北京三星通信技术研究有限公司 图像处理方法、个性化人体显示方法及其图像处理系统
CN103268158B (zh) * 2013-05-21 2017-09-08 上海速盟信息技术有限公司 一种模拟重力传感数据的方法、装置及一种电子设备
JP6136926B2 (ja) * 2013-06-13 2017-05-31 ソニー株式会社 情報処理装置、記憶媒体および情報処理方法
JP2015061579A (ja) * 2013-07-01 2015-04-02 株式会社東芝 動作情報処理装置
JP6433149B2 (ja) 2013-07-30 2018-12-05 キヤノン株式会社 姿勢推定装置、姿勢推定方法およびプログラム
JP6049202B2 (ja) * 2013-10-25 2016-12-21 富士フイルム株式会社 画像処理装置、方法、及びプログラム
US9600887B2 (en) * 2013-12-09 2017-03-21 Intel Corporation Techniques for disparity estimation using camera arrays for high dynamic range imaging
JP6091407B2 (ja) * 2013-12-18 2017-03-08 三菱電機株式会社 ジェスチャ登録装置
KR101700817B1 (ko) 2014-01-10 2017-02-13 한국전자통신연구원 3차원 영상을 이용한 다수의 팔 및 손 검출 장치 및 그 방법
JP6353660B2 (ja) 2014-02-06 2018-07-04 日本放送協会 手話単語分類情報生成装置およびそのプログラム
JP6311372B2 (ja) * 2014-03-13 2018-04-18 オムロン株式会社 画像処理装置および画像処理方法
US10504231B2 (en) * 2014-05-21 2019-12-10 Millennium Three Technologies, Inc. Fiducial marker patterns, their automatic detection in images, and applications thereof
US10426372B2 (en) * 2014-07-23 2019-10-01 Sony Corporation Image registration system with non-rigid registration and method of operation thereof
WO2016090605A1 (en) * 2014-12-11 2016-06-16 Intel Corporation Avatar selection mechanism
CN107533356A (zh) 2015-03-09 2018-01-02 文塔纳3D有限责任公司 头像控制系统
CN104700433B (zh) 2015-03-24 2016-04-27 中国人民解放军国防科学技术大学 一种基于视觉的实时人体全身体运动捕捉方法及其系统
US10022628B1 (en) 2015-03-31 2018-07-17 Electronic Arts Inc. System for feature-based motion adaptation
CN104866101B (zh) * 2015-05-27 2018-04-27 世优（北京）科技有限公司 虚拟对象的实时互动控制方法及装置
US10430867B2 (en) * 2015-08-07 2019-10-01 SelfieStyler, Inc. Virtual garment carousel
DE102015215513A1 (de) * 2015-08-13 2017-02-16 Avl List Gmbh System zur Überwachung einer technischen Vorrichtung
JP2017138915A (ja) * 2016-02-05 2017-08-10 株式会社バンダイナムコエンターテインメント 画像生成システム及びプログラム
US9460557B1 (en) 2016-03-07 2016-10-04 Bao Tran Systems and methods for footwear fitting
JP6723061B2 (ja) * 2016-04-15 2020-07-15 キヤノン株式会社 情報処理装置、情報処理装置の制御方法およびプログラム
CN106023288B (zh) * 2016-05-18 2019-11-15 浙江大学 一种基于图像的动态替身构造方法
CN106296778B (zh) * 2016-07-29 2019-11-15 网易（杭州）网络有限公司 虚拟对象运动控制方法与装置
CN106251396B (zh) 2016-07-29 2021-08-13 迈吉客科技（北京）有限公司 三维模型的实时控制方法和系统
CA3034639C (en) * 2016-09-16 2021-03-23 Verb Surgical Inc. Robotic arms
WO2018069981A1 (ja) * 2016-10-11 2018-04-19 富士通株式会社 運動認識装置、運動認識プログラムおよび運動認識方法
CN108229239B (zh) 2016-12-09 2020-07-10 武汉斗鱼网络科技有限公司 一种图像处理的方法及装置
KR101867991B1 (ko) 2016-12-13 2018-06-20 한국과학기술원 다관절 오브젝트의 모션 에디팅 방법 및 장치
CN106920274B (zh) * 2017-01-20 2020-09-04 南京开为网络科技有限公司 移动端2d关键点快速转换为3d融合变形的人脸建模方法
JP2018119833A (ja) * 2017-01-24 2018-08-02 キヤノン株式会社 情報処理装置、システム、推定方法、コンピュータプログラム、及び記憶媒体
JP2018169720A (ja) * 2017-03-29 2018-11-01 富士通株式会社 動き検出システム
CN107272884A (zh) * 2017-05-09 2017-10-20 聂懋远 一种基于虚拟现实技术的控制方法及其控制系统
CN107154069B (zh) * 2017-05-11 2021-02-02 上海微漫网络科技有限公司 一种基于虚拟角色的数据处理方法及系统
CN107220933B (zh) * 2017-05-11 2021-09-21 上海联影医疗科技股份有限公司 一种参考线确定方法和系统
WO2018207388A1 (ja) 2017-05-12 2018-11-15 ブレイン株式会社 モーションキャプチャに関するプログラム、装置および方法
CN108876879B (zh) * 2017-05-12 2022-06-14 腾讯科技（深圳）有限公司 人脸动画实现的方法、装置、计算机设备及存储介质
US10379613B2 (en) 2017-05-16 2019-08-13 Finch Technologies Ltd. Tracking arm movements to generate inputs for computer systems
CN107578462A (zh) * 2017-09-12 2018-01-12 北京城市系统工程研究中心 一种基于实时运动捕捉的骨骼动画数据处理方法
CN108205654B (zh) * 2017-09-30 2021-06-04 北京市商汤科技开发有限公司 一种基于视频的动作检测方法及装置
CN108229332B (zh) * 2017-12-08 2020-02-14 华为技术有限公司 骨骼姿态确定方法、装置及计算机可读存储介质
CN107958479A (zh) * 2017-12-26 2018-04-24 南京开为网络科技有限公司 一种移动端3d人脸增强现实实现方法
CN108062783A (zh) * 2018-01-12 2018-05-22 北京蜜枝科技有限公司 面部动画映射系统及方法
CN108227931A (zh) * 2018-01-23 2018-06-29 北京市商汤科技开发有限公司 用于控制虚拟人物的方法、设备、系统、程序和存储介质
CN108357595B (zh) * 2018-01-26 2020-11-20 浙江大学 一种基于模型的自平衡无人驾驶自行车及其模型驱动的控制方法
CN108305321B (zh) * 2018-02-11 2022-09-30 牧星天佑(北京)科技文化发展有限公司 一种基于双目彩色成像系统的立体人手3d骨架模型实时重建方法和装置
CN108364254B (zh) 2018-03-20 2021-07-23 北京奇虎科技有限公司 图像处理方法、装置及电子设备
JP6973258B2 (ja) * 2018-04-13 2021-11-24 オムロン株式会社 画像解析装置、方法およびプログラム
CN108427942A (zh) * 2018-04-22 2018-08-21 广州麦仑信息科技有限公司 一种基于深度学习的手掌检测与关键点定位方法
CN108648280B (zh) * 2018-04-25 2023-03-31 深圳市商汤科技有限公司 虚拟角色驱动方法及装置、电子设备和存储介质
CN108829232B (zh) * 2018-04-26 2021-07-23 深圳市同维通信技术有限公司 基于深度学习的人体骨骼关节点三维坐标的获取方法
CN108765272B (zh) * 2018-05-31 2022-07-08 Oppo广东移动通信有限公司 图像处理方法、装置、电子设备及可读存储介质
CN108830783B (zh) * 2018-05-31 2021-07-02 北京市商汤科技开发有限公司 一种图像处理方法、装置和计算机存储介质
CN108830784A (zh) * 2018-05-31 2018-11-16 北京市商汤科技开发有限公司 一种图像处理方法、装置和计算机存储介质
CN108765274A (zh) * 2018-05-31 2018-11-06 北京市商汤科技开发有限公司 一种图像处理方法、装置和计算机存储介质
CN109035415B (zh) 2018-07-03 2023-05-16 百度在线网络技术（北京）有限公司 虚拟模型的处理方法、装置、设备和计算机可读存储介质
WO2020016963A1 (ja) * 2018-07-18 2020-01-23 日本電気株式会社 情報処理装置、制御方法、及びプログラム
CN109101901B (zh) * 2018-07-23 2020-10-27 北京旷视科技有限公司 人体动作识别及其神经网络生成方法、装置和电子设备
CN109146629B (zh) * 2018-08-16 2020-11-27 连云港伍江数码科技有限公司 目标对象的锁定方法、装置、计算机设备和存储介质
CN109191593A (zh) * 2018-08-27 2019-01-11 百度在线网络技术（北京）有限公司 虚拟三维模型的运动控制方法、装置及设备
CN109325450A (zh) 2018-09-25 2019-02-12 Oppo广东移动通信有限公司 图像处理方法、装置、存储介质及电子设备
US10832472B2 (en) * 2018-10-22 2020-11-10 The Hong Kong Polytechnic University Method and/or system for reconstructing from images a personalized 3D human body model and thereof
CN115631305A (zh) 2018-12-29 2023-01-20 深圳市瑞立视多媒体科技有限公司 一种虚拟人物的骨骼模型的驱动方法、插件及终端设备
CN110139115B (zh) 2019-04-30 2020-06-09 广州虎牙信息科技有限公司 基于关键点的虚拟形象姿态控制方法、装置及电子设备
CN110688008A (zh) 2019-09-27 2020-01-14 贵州小爱机器人科技有限公司 虚拟形象交互方法和装置
CN110889382A (zh) 2019-11-29 2020-03-17 深圳市商汤科技有限公司 虚拟形象渲染方法及装置、电子设备和存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180225842A1 (en) * 2016-01-21 2018-08-09 Tencent Technology (Shenzhen) Company Limited Method and apparatus for determining facial pose angle, and computer storage medium
CN108830200A (zh) * 2018-05-31 2018-11-16 北京市商汤科技开发有限公司 一种图像处理方法、装置和计算机存储介质
CN109117726A (zh) * 2018-07-10 2019-01-01 深圳超多维科技有限公司 一种识别认证方法、装置、系统及存储介质
CN109146769A (zh) * 2018-07-24 2019-01-04 北京市商汤科技开发有限公司 图像处理方法及装置、图像处理设备及存储介质
CN109242789A (zh) * 2018-08-21 2019-01-18 成都旷视金智科技有限公司 图像处理方法、图像处理装置及存储介质
CN109376671A (zh) * 2018-10-30 2019-02-22 北京市商汤科技开发有限公司 图像处理方法、电子设备及计算机可读介质

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114333228A (zh) * 2020-09-30 2022-04-12 北京君正集成电路股份有限公司 一种婴儿的智能视频看护方法
CN114333228B (zh) * 2020-09-30 2023-12-08 北京君正集成电路股份有限公司 一种婴儿的智能视频看护方法
CN115564689A (zh) * 2022-10-08 2023-01-03 上海宇勘科技有限公司 一种基于按块处理技术的人工智能图像处理方法及系统

Also Published As

Publication number Publication date
KR20210011984A (ko) 2021-02-02
CN111460873B (zh) 2024-06-11
CN111460874A (zh) 2020-07-28
KR20210011425A (ko) 2021-02-01
SG11202011595QA (en) 2020-12-30
JP7001841B2 (ja) 2022-01-20
SG11202011600QA (en) 2020-12-30
US11741629B2 (en) 2023-08-29
US20210074005A1 (en) 2021-03-11
SG11202011596WA (en) 2020-12-30
US11538207B2 (en) 2022-12-27
JP2021518023A (ja) 2021-07-29
US20210074006A1 (en) 2021-03-11
KR20210011985A (ko) 2021-02-02
JP2021518960A (ja) 2021-08-05
CN111460875B (zh) 2022-03-01
US20210035344A1 (en) 2021-02-04
CN111460871A (zh) 2020-07-28
CN111460872B (zh) 2024-04-16
CN111460875A (zh) 2020-07-28
CN114399826A (zh) 2022-04-26
US20210074003A1 (en) 2021-03-11
CN111460873A (zh) 2020-07-28
SG11202011599UA (en) 2020-12-30
CN111460871B (zh) 2023-12-22
KR20210011424A (ko) 2021-02-01
JP7109585B2 (ja) 2022-07-29
JP2021525431A (ja) 2021-09-24
JP7061694B2 (ja) 2022-04-28
CN114399826B (zh) 2025-12-12
US20210074004A1 (en) 2021-03-11
CN111460870A (zh) 2020-07-28
JP2021524113A (ja) 2021-09-09
CN111460872A (zh) 2020-07-28
SG11202010399VA (en) 2020-11-27
US11468612B2 (en) 2022-10-11

Similar Documents

Publication Publication Date Title
JP7109585B2 (ja) 画像処理方法及び装置、画像デバイス並びに記憶媒体
US11900557B2 (en) Three-dimensional face model generation method and apparatus, device, and medium
CN111353506B (zh) 自适应的视线估计方法和设备
CN108846793B (zh) 基于图像风格转换模型的图像处理方法和终端设备
US11238569B2 (en) Image processing method and apparatus, image device, and storage medium
WO2020207270A1 (zh) 一种三维人脸重建方法及其系统、装置、存储介质
TWI750710B (zh) 圖像處理方法及裝置、圖像處理設備及儲存媒體
JP2019096113A (ja) キーポイントデータに関する加工装置、方法及びプログラム
WO2018137455A1 (zh) 一种图像互动方法及互动装置
US12586329B2 (en) Modeling method, device, and system for three-dimensional head model, and storage medium
WO2023066120A1 (zh) 图像处理方法、装置、电子设备及存储介质
WO2020186385A1 (zh) 图像处理方法、电子设备及计算机可读存储介质
US11450068B2 (en) Method and device for processing image, and storage medium using 3D model, 2D coordinates, and morphing parameter
CN113706373A (zh) 模型重建方法及相关装置、电子设备和存储介质
TWI743843B (zh) 圖像處理方法、圖像處理設備及儲存介質
CN111476151A (zh) 眼球检测方法、装置、设备及存储介质
CN112507766B (zh) 人脸图像提取方法、存储介质及终端设备
CN110070481B (zh) 用于面部的虚拟物品的图像生成方法、装置、终端及存储介质
CN110852934A (zh) 图像处理方法及装置、图像设备及存储介质
CN111652023B (zh) 嘴型的调整、直播方法、装置、电子设备和存储介质
CN110852933A (zh) 图像处理方法及装置、图像处理设备及存储介质
US20260051145A1 (en) Key-point associating apparatus, key-point associating method, and non-transitory computer-readable storage medium
CN118135656A (zh) 基于超图卷积的三维人体姿态估计方法、装置和电子设备
CN113469903A (zh) 图像处理方法、装置、电子设备及可读存储介质
US12505629B1 (en) Model-based augmented image generation

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2020558530

Country of ref document: JP

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19918682

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19918682

Country of ref document: EP

Kind code of ref document: A1