WO2015188767A1 - 一种获取衣服图像属性点的方法和装置 - Google Patents

一种获取衣服图像属性点的方法和装置 Download PDF

Info

Publication number
WO2015188767A1
WO2015188767A1 PCT/CN2015/081261 CN2015081261W WO2015188767A1 WO 2015188767 A1 WO2015188767 A1 WO 2015188767A1 CN 2015081261 W CN2015081261 W CN 2015081261W WO 2015188767 A1 WO2015188767 A1 WO 2015188767A1
Authority
WO
WIPO (PCT)
Prior art keywords
point
distance
shoulder
clothing image
points
Prior art date
Application number
PCT/CN2015/081261
Other languages
English (en)
French (fr)
Inventor
赵刚
Original Assignee
北京京东尚科信息技术有限公司
北京京东世纪贸易有限公司
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 北京京东尚科信息技术有限公司, 北京京东世纪贸易有限公司 filed Critical 北京京东尚科信息技术有限公司
Priority to US15/318,125 priority Critical patent/US10095952B2/en
Publication of WO2015188767A1 publication Critical patent/WO2015188767A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0641—Shopping interfaces
    • G06Q30/0643—Graphical representation of items or shoppers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Definitions

  • the present invention relates to the field of computer technologies, and in particular, to a method and apparatus for acquiring a clothing image attribute point.
  • the way to obtain the attribute points of the clothes image mainly depends on the manual, that is, the person clicks on the position on the clothes image that needs to be calibrated, and the attribute points are formed at the position.
  • the image of the clothes is generally a front view of the clothes, and the image of the clothes is bilaterally symmetrical.
  • the attribute points of the clothes image are more than a dozen, and the manual calibration method is time-consuming and inefficient.
  • the present invention provides a method and apparatus for acquiring a clothing image attribute point, which can improve the efficiency of acquiring a clothing image attribute point.
  • a method of acquiring a clothing image attribute point is provided.
  • the method for obtaining the clothing image attribute point of the present invention comprises: for the clothes image presented on the front side, storing five attribute points of the artificial calibration: the shoulder point on the left and right sides, the midpoint of the hem, and the midpoint of the sleeve end of the left and right sleeves;
  • the horizontal direction is the abscissa and the vertical direction is the ordinate.
  • the first point is the first side The upper end point; the second point is located at the shoulder edge of the first side, the abscissa is the midpoint of the shoulder point of the first point and the first side; the third point is the first side The outer end of the sleeve end; the fourth point is the inner side of the sleeve end of the first side; the fifth point and the sixth point are respectively located on the outer side edge of the sleeve on the first side and the inner side edge of the sleeve, and the two are connected a perpendicular to the connecting line of the shoulder point of the first side and the third point, the ordinate of the foot is located at a preset position between the shoulder point of the first side and the third point;
  • the seventh point is the hem angle of the first side; the eighth point to the eleventh point are located at the midpoint from the second point and the shoulder point of the first side to the seventh
  • a ratio of the distance from the foot to the third point and the distance from the third point to the shoulder point of the first side is 0.35 to 0.45:1.
  • a ratio of a distance from the eighth point to the second point to a distance from the second point to the seventh point is 0.18 to 0.22:1; and the ninth point to the second point
  • the ratio of the distance from the second point to the seventh point is 0.28 to 0.32:1; the distance from the 10th point to the second point is the distance from the second point to the seventh point
  • the ratio is 0.58 to 0.62:1; the ratio of the distance from the second point to the distance from the second point to the seventh point is 0.88 to 0.92:1.
  • an apparatus for obtaining a clothing image attribute point is provided.
  • the device for obtaining the clothing image attribute point of the present invention comprises: a basic point module for storing five attribute points manually marked for the clothes image presented on the front side: a shoulder point on the left and right sides, a midpoint on the hem, and two sleeves on the left and right sides a midpoint of the sleeve end; a determination module for taking the horizontal direction as the abscissa and the vertical direction as the ordinate, for the left and right symmetrical sides of the clothing image
  • the first side determines the first point to the eleventh point: the first point is the upper end point of the first side; the second point is located at the shoulder edge of the first side, and the horizontal point is marked as The midpoint of the first point and the shoulder point of the first side; the third point is the outer side of the sleeve side of the first side; the fourth point is the inner side of the sleeve side of the first side; a point and a sixth point are respectively located on the outer side edge of the sleeve on the first side
  • a ratio of the distance from the foot to the third point and the distance from the third point to the shoulder point of the first side is 0.35 to 0.45:1.
  • a ratio of a distance from the eighth point to the second point to a distance from the second point to the seventh point is 0.18 to 0.22:1; and the ninth point to the second point
  • the ratio of the distance from the second point to the seventh point is 0.28 to 0.32:1; the distance from the 10th point to the second point is the distance from the second point to the seventh point
  • the ratio is 0.58 to 0.62:1; the ratio of the distance from the second point to the distance from the second point to the seventh point is 0.88 to 0.92:1.
  • the clothes image five points are first manually calibrated, and then the other 11 attribute points can be determined by the computer, which reduces the manual workload and improves the efficiency of obtaining the clothing image attribute points.
  • FIG. 1 is a schematic illustration of five points manually calibrated on a garment image in accordance with an embodiment of the present invention
  • FIG. 2 is a schematic diagram of an automatically acquired clothing image attribute point according to an embodiment of the present invention.
  • FIG. 3 is a basic diagram of an apparatus for acquiring a clothing image attribute point according to an embodiment of the present invention. Schematic diagram of the structure.
  • FIG. 1 is a schematic illustration of five points manually calibrated on a garment image in accordance with an embodiment of the present invention. As shown in FIG. 1, a total of five points P1 to P5 are calibrated on the clothing image 10, which are respectively located at the shoulder point on the left and right sides, the midpoint of the hem, and the midpoint of the sleeve end of the left and right sleeves. For the sake of clarity, the outline of the garment is mainly shown in FIG.
  • FIG. 2 is a schematic diagram of an automatically acquired clothing image attribute point according to an embodiment of the present invention.
  • points A1 to A11 in FIG. 2 are sequentially determined, which are located on the left side of the left and right symmetrical clothing image 10, and then the right side of the clothing image and the point A1 to point A11 are determined according to the left and right symmetry of the clothing image. Symmetrical points can be.
  • the coordinate system shown in the figure can be created with the X axis horizontal to the right and the Y axis vertically downward.
  • point A2 is determined.
  • Point A2 is located at the edge of the shoulder and its abscissa is A1 and P1. midpoint.
  • the points of the edge of the shoulder can be scanned from top to bottom and the abscissa of the scanned point can be checked at the same time, and the point scanned is determined as the point A2 when the abscissa is equal to the A1 abscissa and the P1 abscissa average.
  • Fig. 2 take a point M directly on the line connecting A1 and P1.
  • the abscissa of M point is equal to the average of A1 abscissa and P1 abscissa, and then M is used as a vertical line.
  • the intersection of the edge of the part is point A2. This can omit the step of edge detection.
  • the points A3 and A4 are determined. They are the outer and inner points of the sleeve end, respectively.
  • the sleeve end is the horizontal or slightly inclined line that the cuff edge presents on the image. It can be scanned from the point P4 along the sleeve end to both sides, and the two end points of the scanned sleeve end are points A3 and A4, respectively. It is also possible to scan the pixel from the point P4 to the left in the horizontal direction, scan to one pixel, and then scan up and down respectively from the pixel in the vertical direction, when the adjacent two scans in the same direction are respectively transparent.
  • the opacity point is the cuff edge
  • the opaque point continues to scan in the horizontal direction pixel by pixel, that is, the above process is repeated, that is, the sleeve end outer point A3 is obtained.
  • scanning from the point P4 to the right in the horizontal direction is performed pixel by pixel, that is, the cuff end inner point A4 is obtained.
  • the outer side refers to a side away from the vertical axis of symmetry of the clothes.
  • points A5, A6 can be determined.
  • the N is the perpendicular of the line, and the two sides of the sleeve.
  • the intersection point is point A5, A6. It can be scanned from the point N along the vertical line to the outside and inside of the sleeve, respectively. When scanning to the transparent point, the last opaque point scanned is the point A5, A6.
  • point A7 is determined. Scanning from the beginning of P3 to the left or to the outside is similar to determining point A3.
  • Points A8 to A11 can be determined after the point A7 is determined. First, take the point P1 and the midpoint N of the point A2, and then connect the point N with the point A7, and the points A8 to A11 are distributed on the connecting line. Taking the parameter T, the distance from the point to the point N on the connection line is occupied by the distance from the point N to the point A7, and then A8 to A11 are located at positions of 0.2, 0.3, 0.6, and 0.8, respectively. The above ratio There are minor changes to each.
  • points A1 to A11 After the above points A1 to A11 are determined, the symmetry point on the other side of the clothing image can be obtained according to the left and right symmetry of the clothing image.
  • points A to A11 and point P1 plus the symmetry points of these points on the right side of the clothes image, can basically determine the shape of most clothes images, for the images of complicated clothes,
  • the attribute points can be manually added manually.
  • the apparatus 30 for acquiring a clothing image attribute point in the embodiment of the present invention mainly includes a basic point module 31 and a determining module 32.
  • the basic point module 31 is used to store five attribute points manually marked for the clothes image presented on the front side: the shoulder point on the left and right sides, the midpoint of the hem, and the midpoint of the sleeve end of the left and right sleeves.
  • the determining module 32 is configured to use the horizontal direction as the abscissa and the vertical direction as the ordinate. For the first side of the left and right symmetrical sides of the clothing image, point A1 to point A11 are sequentially determined in the manner described above.
  • the clothes image for the clothes image, five points are first manually calibrated, and then the other 11 attribute points can be determined by the computer, which reduces the manual workload and improves the efficiency of obtaining the clothing image attribute points.
  • the objects of the invention can also be achieved by running a program or a set of programs on any computing device.
  • the computing device can be a well-known general purpose device. Therefore, The object of the present invention can also be achieved merely by providing a program product comprising program code for implementing the method or apparatus. That is to say, such a program product also constitutes the present invention, and a storage medium storing such a program product also constitutes the present invention. It will be apparent that the storage medium may be any known storage medium or any storage medium developed in the future.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Image Analysis (AREA)
  • Details Of Garments (AREA)
  • Processing Or Creating Images (AREA)
  • Image Processing (AREA)

Abstract

一种获取衣服图像属性点的方法和装置，能够提高获取衣服图像属性点的效率。该方法包括：对于正面呈现的衣服图像，保存人工标定的5个属性点：左右两侧的肩角点、下摆中点、左右两袖的袖端中点；以水平方向为横坐标，竖直方向为纵坐标，对于所述衣服图像的左右对称两侧中的第一侧，依次确定第1点至第11点。该技术方案对于衣服图像，首先由人工标定5个点，接下来就可以由计算机确定其他11个属性点，减少了人工工作量，提高了获取衣服图像属性点的效率。

Description

一种获取衣服图像属性点的方法和装置 技术领域
本发明涉及计算机技术领域，特别地涉及一种获取衣服图像属性点的方法和装置。
背景技术
随着电子商务的发展，网购成为越来越多的人的消费方式。而购买服装又是人们的主要消费方向之一，但因服装通常需试穿，于是虚拟试衣技术应运而生。虚拟试衣是指利用计算机技术让虚拟模特代替真实用户试穿网上出售的服装，通过虚拟模特试穿呈现的效果给用户选购网上服装一定的参考，便于用户购买到合适的服装。虚拟试衣中所使用的衣服需要进行一些处理，除了对衣服图像进行一些图像处理外，还需要获取衣服图像上的一些属性点，这些属性点与虚拟试衣模特身上的点相对应，从而使虚拟试衣软件能够准确的将衣服“穿”到虚拟试衣模特身上。
目前获取衣服图像属性点的方式主要依赖人工，即由人员点击衣服图像上需要标定的位置，在该位置形成属性点。衣服图像一般是衣服的正面视图，此时衣服图像呈左右对称。一般衣服图像的属性点要在十几个以上，人工标定的方式比较费时，效率较低。
发明内容
有鉴于此，本发明提供一种获取衣服图像属性点的方法和装置，能够提高获取衣服图像属性点的效率。
为实现上述目的，根据本发明的一个方面，提供了一种获取衣服图像属性点的方法。
本发明的获取衣服图像属性点的方法包括：对于正面呈现的衣服图像，保存人工标定的5个属性点：左右两侧的肩角点、下摆中点、左右两袖的袖端中点；以水平方向为横坐标，竖直方向为纵坐标，对于所述衣服图像的左右对称两侧中的第一侧，依次确定如下的第1点至第11点：第1点为所述第一侧的上方端点；第2点位于所述第一侧的肩部边缘，其横坐标为所述第1点与所述第一侧的肩角点的中点；第3点为所述第一侧的袖端外侧点；第4点为所述第一侧的袖端内侧点；第5点和第6点分别位于所述第一侧的袖中外侧边缘和袖中内侧边缘，二者连线垂直于所述第一侧的肩角点与所述第3点的连接线，垂足的纵坐标位于所述第一侧的肩角点与所述第3点之间的预设位置处；第7点为所述第一侧的下摆角点；第8点至第11点位于从第2点与所述第一侧的肩角点的中点至第7点的连线上，按预设位置分布；确定所述衣服图像的左右对称两侧中的第二侧上的与上述第1点至第11点相对称的点。
可选地，所述垂足至所述第3点的距离与所述第3点至所述第一侧的肩角点距离的比为0.35～0.45∶1。
可选地，所述第8点至所述第2点的距离与所述第2点至所述第7点距离的比为0.18～0.22∶1；所述第9点至所述第2点的距离与所述第2点至所述第7点距离的比为0.28～0.32∶1；所述第10点至所述第2点的距离与所述第2点至所述第7点距离的比为0.58～0.62∶1；所述第11点所述第2点的距离与所述第2点至所述第7点距离的比为0.88～0.92∶1。
根据本发明的另一方面，提供了获取衣服图像属性点的装置。
本发明的获取衣服图像属性点的装置包括：基本点模块，用于对于正面呈现的衣服图像，保存人工标定的5个属性点：左右两侧的肩角点、下摆中点、左右两袖的袖端中点；确定模块，用于以水平方向为横坐标，竖直方向为纵坐标，对于所述衣服图像的左右对称两侧中 的第一侧，依次确定如下的第1点至第11点：第1点为所述第一侧的上方端点；第2点位于所述第一侧的肩部边缘，其横从标为所述第1点与所述第一侧的肩角点的中点；第3点为所述第一侧的袖端外侧点；第4点为所述第一侧的袖端内侧点；第5点和第6点分别位于所述第一侧的袖中外侧边缘和袖中内侧边缘，二者连线垂直于所述第一侧的肩角点与所述第3点的连接线，垂足的纵坐标位于所述第一侧的肩角点与所述第3点之间的预设位置处；第7点为所述第一侧的下摆角点；第8点至第11点位于从第2点与所述第一侧的肩角点的中点至第7点的连线上，按预设位置分布；确定所述衣服图像的左右对称两侧中的第一侧上的与上述第1点至第11点相对称的点。
可选地，所述垂足至所述第3点的距离与所述第3点至所述第一侧的肩角点距离的比为0.35～0.45∶1。
可选地，所述第8点至所述第2点的距离与所述第2点至所述第7点距离的比为0.18～0.22∶1；所述第9点至所述第2点的距离与所述第2点至所述第7点距离的比为0.28～0.32∶1；所述第10点至所述第2点的距离与所述第2点至所述第7点距离的比为0.58～0.62∶1；所述第11点所述第2点的距离与所述第2点至所述第7点距离的比为0.88～0.92∶1。
根据本发明的技术方案，对于衣服图像，首先由人工标定5个点，接下来就可以由计算机确定其他11个属性点，减少了人工工作量，提高了获取衣服图像属性点的效率。
附图说明
附图用于更好地理解本发明，不构成对本发明的不当限定。其中：
图1是根据本发明实施例的在衣服图像上人工标定的5个点的示意图；
图2是根据本发明实施例的自动获取的衣服图像属性点的示意图；
图3是根据本发明实施例的获取衣服图像属性点的装置的基本结 构的示意图。
具体实施方式
以下结合附图对本发明的示范性实施例做出说明，其中包括本发明实施例的各种细节以助于理解，应当将它们认为仅仅是示范性的。因此，本领域普通技术人员应当认识到，可以对这里描述的实施例做出各种改变和修改，而不会背离本发明的范围和精神。同样，为了清楚和简明，以下的描述中省略了对公知功能和结构的描述。
在本发明实施例中，首先对衣服图像人工标定5个点，然后根据这5个点由计算机确定其他点，因为减少了人工标定点的个数，所以提高了获取衣服图像属性点的效率。人工标定的5个点如图1所示。图1是根据本发明实施例的在衣服图像上人工标定的5个点的示意图。如图1所示，在衣服图像10上标定了P1至P5共5个点，分别位于左右两侧的肩角点、下摆中点、以及左右两袖的袖端中点。为示意清晰，图1中主要示出衣服的轮廓。
以下说明在标定了P1至P5之后其他点的获取方式。使用计算机可按该方式自动得出。参考图2，图2是根据本发明实施例的自动获取的衣服图像属性点的示意图。在本实施例中，依次确定图2中的点A1至点A11，它们位于左右对称的衣服图像10的左侧，然后根据衣服图像的左右对称，确定衣服图像右侧的与点A1至点A11对称的点即可。
可以建立图中所示的坐标系，X轴水平向右，Y轴竖直向下。首先确定点A1。可从X轴起沿Y轴方向即从上至下逐像素扫描(以下如无特殊说明，“扫描”都为逐像素扫描)衣服图像，在衣服图像10竖直的对称轴(图中未示出)左侧扫描到第1个不透明点时，将该点确定为点A1。
接下来确定点A2。点A2位于肩部边缘，其横坐标为A1和P1的 中点。可以从上至下扫描肩部的边缘各点并同时检查扫描到的点的横坐标，当该横坐标等于A1横坐标与P1横坐标平均值时将扫描到的点确定为点A2。也可以如图2所示，直接在A1与P1的连线上取一点M，M点横坐标等于A1横坐标与P1横坐标平均值，再过M作竖直线，该竖直线与肩部边缘的交点即为点A2。这样可以省略边缘检测的步骤。
接下来确定点A3、A4。它们分别是袖端的外侧点和内侧点。袖端是袖口边缘在图像上呈现的水平或者略为倾斜的线。可以从点P4起沿着袖端分别向两边扫描，扫描到的袖端的两个端点即分别是点A3、A4。也可以从点P4点起沿水平方向向左逐像素扫描，扫描到一个像素之后从该像素起再沿竖直方向分别向上和向下扫描，当同一方向的相邻的两次扫描分别为透明点和不透明点时，说明该不透明点为袖口边缘，自该不透明点继续沿水平方向逐像素扫描，即重复上述过程，即得到袖端外侧点A3。按上述方式，从点P4点起沿水平方向向右逐像素扫描，即得到袖端内侧点A4。本实施例的描述中，外侧是指远离衣服竖直对称轴的一侧。
在确定A3之后，可以确定点A5、A6。此时可以先连接P1和A3，在连线上取一点N，N与A3的距离大致为P1与A3距离的0.35至0.45倍，过N作该连线的垂线，与袖子边缘的两个交点即为点A5、A6。可以从点N起沿着该垂线分别向袖子外侧和内侧扫描，当扫描到透明点时，最后一次扫描到的不透明点即为点A5、A6。
接下来确定点A7。从P3开始向左即向外侧扫描，方式与确定点A3相类似。
在确定点A7之后就可以确定点A8至A11。先取点P1与点A2的中点N，再连接点N与点A7，点A8至A11即分布在连接线上。取参数T，表示该连接线上的点至点N的距离占点N至点A7距离的占比，则A8至A11分别位于占比为0.2、0.3、0.6、0.8的位置处。上述占比 各允许有小幅更改。
在确定上述的点A1至A11之后，根据衣服图像的左右对称性可以得到衣服图像另一侧的对称点。在虚拟试衣的实现过程中，点A至A11以及点P1，加上这些点在衣服图像的右侧的对称点，能够基本确定绝大部分衣服图像的形状，对于设计复杂的衣服的图像，可以再由人工手动增加属性点。
图3是根据本发明实施例的获取衣服图像属性点的装置的基本结构的示意图。该装置可作为一种软件设置在计算机中。如图3所示，本发明实施例的获取衣服图像属性点的装置30主要包括基本点模块31和确定模块32。基本点模块31用于对于正面呈现的衣服图像，保存人工标定的5个属性点：左右两侧的肩角点、下摆中点、左右两袖的袖端中点。确定模块32用于以水平方向为横坐标，竖直方向为纵坐标，对于衣服图像的左右对称两侧中的第一侧，按上文描述的方式依次确定点A1至点A11。
根据本发明实施例的技术方案，对于衣服图像，首先由人工标定5个点，接下来就可以由计算机确定其他11个属性点，减少了人工工作量，提高了获取衣服图像属性点的效率。
以上结合具体实施例描述了本发明的基本原理，但是，需要指出的是，对本领域的普通技术人员而言，能够理解本发明的方法和设备的全部或者任何步骤或者部件，可以在任何计算装置(包括处理器、存储介质等)或者计算装置的网络中，以硬件、固件、软件或者它们的组合加以实现，这是本领域普通技术人员在阅读了本发明的说明的情况下运用他们的基本编程技能就能实现的。
因此，本发明的目的还可以通过在任何计算装置上运行一个程序或者一组程序来实现。所述计算装置可以是公知的通用装置。因此， 本发明的目的也可以仅仅通过提供包含实现所述方法或者装置的程序代码的程序产品来实现。也就是说，这样的程序产品也构成本发明，并且存储有这样的程序产品的存储介质也构成本发明。显然，所述存储介质可以是任何公知的存储介质或者将来开发出的任何存储介质。
还需要指出的是，在本发明的装置和方法中，显然，各部件或各步骤是可以分解和/或重新组合的。这些分解和/或重新组合应视为本发明的等效方案。并且，执行上述系列处理的步骤可以自然地按照说明的顺序按时间顺序执行，但是并不需要一定按照时间顺序执行。某些步骤可以并行或彼此独立地执行。
上述具体实施方式，并不构成对本发明保护范围的限制。本领域技术人员应该明白的是，取决于设计要求和其他因素，可以发生各种各样的修改、组合、子组合和替代。任何在本发明的精神和原则之内所作的修改、等同替换和改进等，均应包含在本发明保护范围之内。

Claims (6)

  1. 一种获取衣服图像属性点的方法，其特征在于，包括：
    对于正面呈现的衣服图像，保存人工标定的5个属性点：左右两侧的肩角点、下摆中点、左右两袖的袖端中点；
    以水平方向为横坐标，竖直方向为纵坐标，对于所述衣服图像的左右对称两侧中的第一侧，依次确定如下的第1点至第11点：
    第1点为所述第一侧的上方端点；
    第2点位于所述第一侧的肩部边缘，其横坐标为所述第1点与所述第一侧的肩角点的中点；
    第3点为所述第一侧的袖端外侧点；
    第4点为所述第一侧的袖端内侧点；
    第5点和第6点分别位于所述第一侧的袖中外侧边缘和袖中内侧边缘，二者连线垂直于所述第一侧的肩角点与所述第3点的连接线，垂足的纵坐标位于所述第一侧的肩角点与所述第3点之间的预设位置处；
    第7点为所述第一侧的下摆角点；
    第8点至第11点位于从第2点与所述第一侧的肩角点的中点至第7点的连线上，按预设位置分布；
    确定所述衣服图像的左右对称两侧中的第二侧上的与上述第1点至第11点相对称的点。
  2. 根据权利要求1所述的方法，其特征在于，
    所述垂足至所述第3点的距离与所述第3点至所述第一侧的肩角点距离的比为0.35～0.45∶1。
  3. 根据权利要求1或2所述的方法，其特征在于，
    所述第8点至所述第2点的距离与所述第2点至所述第7点距离的比为0.18～0.22∶1；
    所述第9点至所述第2点的距离与所述第2点至所述第7点距离的比为0.28～0.32∶1；
    所述第10点至所述第2点的距离与所述第2点至所述第7点距离的比为0.58～0.62∶1；
    所述第11点所述第2点的距离与所述第2点至所述第7点距离的比为0.88～0.92∶1。
  4. 一种获取衣服图像属性点的装置，其特征在于，包括：
    基本点模块，用于对于正面呈现的衣服图像，保存人工标定的5个属性点：左右两侧的肩角点、下摆中点、左右两袖的袖端中点；
    确定模块，用于以水平方向为横坐标，竖直方向为纵坐标，对于所述衣服图像的左右对称两侧中的第一侧，依次确定如下的第1点至第11点：
    第1点为所述第一侧的上方端点；
    第2点位于所述第一侧的肩部边缘，其横坐标为所述第1点与所述第一侧的肩角点的中点；
    第3点为所述第一侧的袖端外侧点；
    第4点为所述第一侧的袖端内侧点；
    第5点和第6点分别位于所述第一侧的袖中外侧边缘和袖中内侧边缘，二者连线垂直于所述第一侧的肩角点与所述第3点的连接线，垂足的纵坐标位于所述第一侧的肩角点与所述第3点之间的预设位置处；
    第7点为所述第一侧的下摆角点；
    第8点至第11点位于从第2点与所述第一侧的肩角点的中点至第7点的连线上，按预设位置分布；
    确定所述衣服图像的左右对称两侧中的第二侧上的与上述第1点至第11点相对称的点。
  5. 根据权利要求4所述的装置，其特征在于，
    所述垂足至所述第3点的距离与所述第3点至所述第一侧的肩角 点距离的比为0.35～0.45∶1。
  6. 根据权利要求4或5所述的装置，其特征在于，
    所述第8点至所述第2点的距离与所述第2点至所述第7点距离的比为0.18～0.22∶1；
    所述第9点至所述第2点的距离与所述第2点至所述第7点距离的比为0.28～0.32∶1；
    所述第10点至所述第2点的距离与所述第2点至所述第7点距离的比为0.58～0.62∶1；
    所述第11点所述第2点的距离与所述第2点至所述第7点距离的比为0.88～0.92∶1。
PCT/CN2015/081261 2014-06-13 2015-06-11 一种获取衣服图像属性点的方法和装置 WO2015188767A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/318,125 US10095952B2 (en) 2014-06-13 2015-06-11 Method and device for acquiring clothing image attribute points

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410264313.8 2014-06-13
CN201410264313.8A CN104008501B (zh) 2014-06-13 2014-06-13 一种获取衣服图像属性点的方法和装置

Publications (1)

Publication Number Publication Date
WO2015188767A1 true WO2015188767A1 (zh) 2015-12-17

Family

ID=51369147

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/081261 WO2015188767A1 (zh) 2014-06-13 2015-06-11 一种获取衣服图像属性点的方法和装置

Country Status (4)

Country Link
US (1) US10095952B2 (zh)
CN (1) CN104008501B (zh)
TW (1) TWI541734B (zh)
WO (1) WO2015188767A1 (zh)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104008501B (zh) 2014-06-13 2017-11-14 北京京东尚科信息技术有限公司 一种获取衣服图像属性点的方法和装置
CN104766369A (zh) * 2015-04-23 2015-07-08 上海趣搭网络科技有限公司 建立服饰模型的方法和装置
WO2017203663A1 (ja) * 2016-05-26 2017-11-30 楽天株式会社 形状弁別装置、形状弁別方法及び形状弁別プログラム
WO2018000048A1 (en) * 2016-06-30 2018-01-04 776BC International Pty Ltd Garments, systems and methods for sports training
US10991115B2 (en) * 2017-03-06 2021-04-27 Rakuten, Inc. Image processing device, image processing method, server, and computer program
CN107742123B (zh) * 2017-10-17 2019-12-03 华南师范大学 基于角点的服装款式特征提取方法及服装检索方法和***
CN110378959A (zh) * 2019-07-15 2019-10-25 杭州恢弘科技有限公司 一种服装辅助印烫定位设置方法、定位方法及辅助印烫方法
CN110672014B (zh) * 2019-08-27 2023-08-22 东莞市精致自动化科技有限公司 一种衣服尺寸测量方法
CN112529670B (zh) * 2021-02-07 2021-04-30 上海英立视电子有限公司 一种虚拟试衣方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070000007A1 (en) * 2005-06-20 2007-01-04 Nike, Inc. Predictive and counter predictive visual stimuli
CN101493930A (zh) * 2008-01-21 2009-07-29 保定市天河电子技术有限公司 一种换装方法和换发方法
CN102682211A (zh) * 2012-05-09 2012-09-19 晨星软件研发(深圳)有限公司 一种立体试衣方法及装置
CN103578004A (zh) * 2013-11-15 2014-02-12 西安工程大学 一种用于展示虚拟试衣效果的方法
CN104008501A (zh) * 2014-06-13 2014-08-27 北京京东尚科信息技术有限公司 一种获取衣服图像属性点的方法和装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20120031279A (ko) * 2012-02-01 2012-04-02 오정철 인터넷을 이용한 의류 판매 방법
US10339593B2 (en) * 2015-07-07 2019-07-02 Lutzy Inc. System and network for outfit planning and wardrobe management
US9996763B2 (en) * 2015-09-18 2018-06-12 Xiaofeng Han Systems and methods for evaluating suitability of an article for an individual

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070000007A1 (en) * 2005-06-20 2007-01-04 Nike, Inc. Predictive and counter predictive visual stimuli
CN101493930A (zh) * 2008-01-21 2009-07-29 保定市天河电子技术有限公司 一种换装方法和换发方法
CN102682211A (zh) * 2012-05-09 2012-09-19 晨星软件研发(深圳)有限公司 一种立体试衣方法及装置
CN103578004A (zh) * 2013-11-15 2014-02-12 西安工程大学 一种用于展示虚拟试衣效果的方法
CN104008501A (zh) * 2014-06-13 2014-08-27 北京京东尚科信息技术有限公司 一种获取衣服图像属性点的方法和装置

Also Published As

Publication number Publication date
US10095952B2 (en) 2018-10-09
TWI541734B (zh) 2016-07-11
CN104008501A (zh) 2014-08-27
TW201546724A (zh) 2015-12-16
CN104008501B (zh) 2017-11-14
US20170124420A1 (en) 2017-05-04

Similar Documents

Publication Publication Date Title
WO2015188767A1 (zh) 一种获取衣服图像属性点的方法和装置
KR102346320B1 (ko) 고속 3d 모델 피팅 및 인체 측정법
US20230014804A1 (en) Processing user selectable product images and facilitating visualization-assisted virtual dressing
US10832472B2 (en) Method and/or system for reconstructing from images a personalized 3D human body model and thereof
US11640672B2 (en) Method and system for wireless ultra-low footprint body scanning
Zhang et al. Quality dynamic human body modeling using a single low-cost depth camera
US20190244407A1 (en) System, device, and method of virtual dressing utilizing image processing, machine learning, and computer vision
CN107292948B (zh) 一种人体建模方法、装置及电子设备
CN110503681B (zh) 人体模型自动创建方法及三维试衣***
CN105279787A (zh) 基于拍照的户型图识别生成三维房型的方法
CN106204718A (zh) 一种基于单个Kinect的简易高效三维人体重建方法
US11922593B2 (en) Methods of estimating a bare body shape from a concealed scan of the body
CN106127773A (zh) 一种基于图片的脚型数据获取方法
CN113469929B (zh) 训练数据生成方法、装置、电子设备及计算机可读存储介质
WO2015165420A1 (zh) 一种处理上衣图像的方法和装置
CN109615585A (zh) 人体图像的姿态矫正方法和装置、电子设备、存储介质
WO2024149158A1 (zh) 数据拼接方法、三维扫描***、电子装置和存储介质
JP2015092331A (ja) 画像処理装置、画像処理方法及びプログラム
He et al. T $^ 3$ Bench: Benchmarking Current Progress in Text-to-3D Generation
US20210241480A1 (en) Measuring Surface Distances on Human Bodies
WO2018182938A1 (en) Method and system for wireless ultra-low footprint body scanning
Sobhiyeh et al. Hole filling in 3D scans for digital anthropometric applications
Senanayake et al. Automated human body measurement extraction: single digital camera (webcam) method–phase 1
CN104933746B (zh) 一种为平面图片设置动态阴影的方法及装置
CN105303380A (zh) 虚拟试衣***

Legal Events

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

Ref document number: 15806860

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 15318125

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 13/04/17)

122 Ep: pct application non-entry in european phase

Ref document number: 15806860

Country of ref document: EP

Kind code of ref document: A1