返回项目页面
对比文件列表
1999-09-28_US5960099A_发明授权_US5960099A System and method for creating a digitized likeness of persons_+++A_B_H_J+++.docx
2026-06-17 01:49
2001-09-04_US6283858B_发明授权_US06283858B1 Method for manipulating images_+++A_B_h_i_j_n_o_p+++.docx
2026-06-17 01:49
2003-04-03_US2003065255A_发明申请_US20030065255A1 Simulation of an aesthetic feature on a facial image_+++A_B_H_J_M_N_P_o+++.docx
2026-06-17 01:49
2003-12-04_US2003223622A_发明申请_US20030223622A1 Method and system for enhancing portrait images_+++A_B_H_J_N_P_l_m_o+++.docx
2026-06-17 01:49
2006-02-08_CN1731416A_发明公开_CN1731416A 快速且精确的人脸特征点定位方法_+++H_J_a_b+++.docx
2026-06-17 01:49
2007-03-08_US2007052726A_发明申请_US20070052726A1 Method and system for likeness reconstruction_+++A_B_N_P_j_m+++.docx
2026-06-17 01:49
2008-08-28_JP2008198189A_发明专利_JP2008198189A Method, system and program for warning monitored user of behavior of monitoring object user showing adverse behavior_+++A_B_H_J+++.docx
2026-06-17 01:49
2009-08-06_US2009196475A_发明申请_US20090196475A1 AUTOMATIC MASK DESIGN AND REGISTRATION AND FEATURE DETECTION FOR COMPUTER-AIDED SKIN ANALYSIS_+++a_b_h_j+++.docx
2026-06-17 01:49
2012-06-07_US2012140042A_发明申请_US20120140042A1 WARNING A USER ABOUT ADVERSE BEHAVIORS OF OTHERS WITHIN AN ENVIRONMENT BASED ON A 3D CAPTURED IMAGE STREAM_+++A_B_h_j+++.docx
2026-06-17 01:49
2012-10-04_WO2012129727A_发明申请_WO2012129727A1 METHOD OF FACIAL LANDMARK DETECTION_+++A_B_H_J+++.docx
2026-06-17 01:49
2012-11-29_US2012299945A_发明申请_US20120299945A1 METHOD, SYSTEM AND COMPUTER PROGRAM PRODUCT FOR AUTOMATIC AND SEMI-AUTOMATIC MODIFICATOIN OF DIGITAL IMAGES OF FACES_+++A_B_E_F_H_I_J_L_M_N_O_P_g+++.docx
2026-06-17 01:49
2013-02-21_US2013044947A_发明申请_US20130044947A1 Methods and Apparatus for Automated Portrait Retouching Using Facial Feature Localization_+++A_B_E_F_H_I_J_N_O_P+++.docx
2026-06-17 01:49
2013-02-21_US2013044958A_发明申请_US20130044958A1 Methods and Apparatus for Automated Facial Feature Localization_+++A_B_D_E_F_H_I_J_N_O_P_R_T+++.docx
2026-06-17 01:49
2013-02-21_US2013044947A_发明申请_US20130044947A1 Methods and Apparatus for Automated Portrait Retouching Using Facial Feature Localization_+++A_B_E_F_H_I_J_N_O_P+++.docx

对比文件的名称:2013-02-21_US2013044947A_发明申请_US20130044947A1 Methods and Apparatus for Automated Portrait Retouching Using Facial Feature Localization

目标专利的名称:1036一种装置和相关联的方法CN105229673B

本次调用的模型名称:资深专利代理师

### 特征比对表格

技术特征描述及公开性判断结果对比文件原文引用公开性论述
技术特征A:包括至少一个处理器——《直接公开》[0124] “In various embodiments, computer system 1000 may be a uniprocessor system including one processor 1010, or a multiprocessor system including several processors 1010 (e.g., two, four, eight, or another suitable number).”对比文件明确公开了计算机系统包括一个或多个处理器(processor 1010),用于执行程序指令以进行面部特征定位和修饰,与特征A中“至少一个处理器”完全相同,因此直接公开。
技术特征B:和至少一个包括计算机程序代码的存储器——《直接公开》[0126] “System memory 1020 may be configured to store program instructions and/or data accessible by processor 1010.” [0073] “The facial retouching module may in some embodiments be implemented by a non-transitory, computer-readable storage medium and one or more processors ... The computer-readable storage medium may store program instructions executable by the one or more processors...”对比文件公开了系统存储器(system memory 1020)存储程序指令(program instructions)和数据,且程序指令由处理器执行,这对应特征B中的“至少一个包括计算机程序代码的存储器”,因此直接公开。
技术特征C:通过检测由用户所指向的现实世界中的脸部或者脸部的现实世界中的图像的位置,来检测用户对与脸部相关联的脸部特征的位置指示,使得与脸部相关联的所述脸部特征与现实世界中的脸部或者脸部的现实世界中的图像相关联——不公开[0051] “In some embodiments, user interactivity is provided as an interface allowing user to move fitted points to a desired place.” [0056] “...if the eye corner is located incorrectly by the embodiment, the user can pull the eye corner to the correct location...” [0107] “...receiving user input representing corrections to selected ones of the feature point locations...”对比文件公开的用户交互是允许用户在计算机生成的图像上移动或校正已检测的特征点位置(通过鼠标、触摸屏等对二维图像的拖拽操作),并未公开检测用户对现实世界中的脸部(如用户自己的真实脸部)或现实世界中的脸部图像(如照片)进行指向的动作。用户交互的对象是计算机生成的图像中的特征点,而非对现实世界物理对象的指向检测,因此特征C未被直接或隐含公开。
技术特征D:基于所检测到的用户对与脸部相关联的脸部特征的位置指示,提供相对应的计算机所生成的脸部特征的位置的锚定,使得相对应的所述计算机所生成的脸部特征的脸部标记定位能够被锚定在计算机所生成的脸部图像上的相对应位置周围。——不公开[0051] “...user interactivity is provided as an interface allowing user to move fitted points to a desired place. Previously-user-moved points are considered to be trusted and are not moved in the next round of iteration.” [0107] “...receiving user input representing corrections to selected ones of the feature point locations, and iterating the applying the profile model and the applying the global and component-based shape models until a change in the feature point locations is less than a threshold, with respect to which the iterating ... further includes treating the corrections to the selected ones of the feature point locations as fixed locations for the selected ones of the feature point locations.”对比文件虽然公开了基于用户校正输入来“固定”特征点位置(即用户移动后的点被视为可信且在下一次迭代中不被移动),但这本质上是基于用户对图像上特征点的拖拽操作。而特征D明确要求“基于所检测到的用户对与脸部相关联的脸部特征的位置指示”,该位置指示必须是通过检测用户对现实世界中的脸部或现实世界中的图像的指向而产生的(如特征C所定义)。由于对比文件未公开特征C的检测方式,因此特征D的锚定基础和方式均与对比文件不同,未被直接或隐含公开。
技术特征E:其中所述装置被配置为执行被锚定在所述计算机所生成的脸部图像上的所述相对应位置周围的、针对相对应脸部特征的脸部标记定位。——《直接公开》[0111] “A global shape model is applied to obtain global shape parameters (block 250). Component shape models are applied to determine local shape parameters and local feature point coordinates for each component (block 255).” [0107] “...and iterating the applying the profile model and the applying the global and component-based shape models until a change in the feature point locations is less than a threshold, with respect to which the iterating the applying the profile model and the applying the global and component-based shape models further includes treating the corrections to the selected ones of the feature point locations as fixed locations for the selected ones of the feature point locations.”对比文件公开了在用户提供校正(即提供锚定点)后,通过迭代应用全局和组件形状模型来执行特征点定位(即脸部标记定位),以确定其他点的位置。这直接对应于在锚定位置周围执行脸部标记定位,因此直接公开。
技术特征F:其中所述装置被配置为通过以下操作来提供所述锚定:调节已经使用脸部标记定位被标识的相对应脸部特征的位置,以将所述位置锚定在所述计算机所生成的脸部图像上的所述相对应位置周围。——《直接公开》[0051] “...user interactivity is provided as an interface allowing user to move fitted points to a desired place. Previously-user-moved points are considered to be trusted and are not moved in the next round of iteration.” [0107] “...receiving user input representing corrections to selected ones of the feature point locations...”对比文件公开了用户移动(调节)已通过脸部标记定位标识的特征点(如眼角)的位置,并将这些用户校正后的点视为固定位置(即锚定)。这直接对应于通过调节已标识特征的位置来提供锚定,因此直接公开。
技术特征G:其中所述装置被配置为通过以下操作来提供所述锚定:针对通过脸部标记定位的首次使用而关联相对应脸部特征的所述位置,使得所述相对应脸部特征的脸部标记定位被锚定在所述计算机所生成的脸部图像上的相对应位置周围。——不公开[0034] “For a new image (not in the training set), the face detection window is used to predict the initial locations of the feature points. Referring briefly to FIG. 2B, the search then proceeds iteratively (e.g., through blocks 250, 245, 250, 255, 260).” [0051] “Previously-user-moved points are considered to be trusted and are not moved in the next round of iteration.”对比文件公开的脸部标记定位流程是:先通过模型预测初始位置,然后迭代搜索和拟合,用户交互是在此过程中或之后对结果进行校正。其首次使用(初始定位)是基于模型预测,而非用户指示的位置。特征G要求首次使用就关联用户指示的位置(即锚定),这与对比文件的先自动定位再校正的顺序不同,因此未被公开。
技术特征H:其中针对相对应脸部特征的所述脸部标记定位包括在所述计算机所生成的脸部图像上定位多个脸部标记点以提供所述特征的定位。——《直接公开》[0030] “Some embodiments employ a face model consisting of seven 7 components: jawline, nose, lips, left eye, right eye, left brow, eye brow. An example of such a model on a face-fitting result of one embodiment is shown in FIG. 4.” [0115] “In input image 400, feature points (circles) are provided for left brow 410, right brow 420, right eye 430, nose 440, lips 450, jawline 460, and right eye 470.”对比文件明确公开了在图像上定位多个特征点(如左眉、右眉、右眼、鼻子、嘴唇、下巴线等)来提供特征的定位,这直接对应于特征H中的“定位多个脸部标记点”,因此直接公开。
技术特征I:其中所述装置被配置为通过调节位于所述计算机所生成的脸部图像上的多个脸部标记点中的一个或多个脸部标记点的位置而提供相对应脸部特征的所述位置的锚定。——《直接公开》[0056] “...if the eye corner is located incorrectly by the embodiment, the user can pull the eye corner to the correct location, and the other point locations are moved according to this user-provided constraint...” [0051] “...user interactivity is provided as an interface allowing user to move fitted points to a desired place.”对比文件公开了用户通过移动(调节)一个或多个特征点(如眼角)的位置来提供校正(锚定),其他点根据约束相应移动。这直接对应于通过调节多个脸部标记点中的一个或多个的位置来提供锚定,因此直接公开。
技术特征J:其中所述多个脸部标记点对应于以下相对应脸部特征中的一个或多个:左眼、右眼、左侧眉毛、右侧眉毛、左侧脸颊、右侧脸颊、脸部轮廓、左耳、右耳、嘴唇、鼻子、前额、脸颊和下巴。——《直接公开》[0030] “...a face model consisting of seven 7 components: jawline, nose, lips, left eye, right eye, left brow, eye brow.” [0115] “...feature points (circles) are provided for left brow 410, right brow 420, right eye 430, nose 440, lips 450, jawline 460, and right eye 470.” [0059] “Eyes: all of the eyes... Brows... Inner mouth... Lips... Skin...”对比文件公开的特征点包括左眼、右眼、眉毛、鼻子、嘴唇、下巴线(脸部轮廓的一部分)等。虽然未明确列出前额、脸颊、耳朵所有点,但[0059]中的皮肤掩膜(skin mask)涉及额头、脸颊区域,且特征J是开放式的“一个或多个”,对比文件公开了其中的主要部分,因此直接公开。
技术特征K:其中所检测到的用户对与脸部相关联的脸部特征的位置指示包括对用户在所述脸部上指向以下之一的检测:左眼、右眼、左侧脸颊、右侧脸颊、左耳、右耳、嘴唇、鼻子、前额或下巴。——不公开[0056] “...if the eye corner is located incorrectly by the embodiment, the user can pull the eye corner to the correct location...”对比文件公开的用户校正针对的是图像上的特征点(如眼角),并未公开检测用户对现实世界中的脸部(或脸部图像)进行指向的动作。虽然用户校正的特征可能包括眼睛、嘴唇等,但检测方式完全不同(是图像拖拽而非现实世界指向),因此特征K未被公开。
技术特征L:其中用户所指示的位置基于在所述用户的位置指示之前或之后对预定义脸部特征的用户选择。——不公开未发现相关内容。对比文件未公开用户在指示位置之前或之后从预定义列表中选择脸部特征的操作。其用户交互是直接对特征点进行拖拽校正,而非先选择特征类型再指示位置,因此未被公开。
技术特征M:其中所述预定义脸部特征是以下之一:左眼、右眼、左侧眉毛、右侧眉毛、左侧脸颊、右侧脸颊、左耳、右耳、嘴唇、鼻子、前额或下巴。——不公开未发现相关内容。基于特征L的判断,对比文件未公开预定义脸部特征的选择,因此特征M也未公开。
技术特征N:其中所述装置被配置为向相对应的脸部特征应用视觉效果。——《直接公开》[0054] “Once the landmarks have been located, masks are constructed for the relevant face regions ... to enable the editing step. Embodiments then present an interactive interface for the user to do common portrait corrections such as skin softening, eye contrast and sharpening, and lip and teeth enhancement.” [0072] “An example of the result of applying these operations is shown in FIG. 3B...”对比文件公开了在定位特征点并生成掩膜后,对相应脸部区域(如皮肤、眼睛、嘴唇)应用修饰效果(如皮肤柔化、眼睛锐化、嘴唇增强等),这直接对应于向脸部特征应用视觉效果,因此直接公开。
技术特征O:其中所述装置被配置为向通过所述脸部标记定位而被定位在所述脸部上的一个或多个脸部标记点所指示的区域应用视觉效果。——《直接公开》[0054] “Once the landmarks have been located, masks are constructed for the relevant face regions (eyes, lips, skin, teeth, and brows) to enable the editing step.” [0099] “...adjusting one or more facial component shapes based on the plurality of masks. ... adjusting one or more facial component colors based on the plurality of masks.”对比文件公开了利用脸部标记点生成掩膜(masks),并基于这些掩膜对相应区域(如眼部、唇部、皮肤区域)应用调整效果(如形状、颜色调整)。这直接对应于向脸部标记点所指示的区域应用视觉效果,因此直接公开。
技术特征P:其中所应用的视觉效果是以下之一:唇膏应用、眼影应用、眼线笔应用、睫毛颜色应用、眉毛颜色应用、脸颊颜色应用、眼睛着色、红眼消除、皮肤纹理平滑、皮肤闪光去除和皮肤瑕疵去除。——《直接公开》[0055] “...provide simple, intuitive, controls to apply the adjustments. Examples of the controls include, skin smoothness, eye sharpness and contrast, whitening of sclera and teeth, lip color and contrast adjustment, skin tone adjustment, and control of lighting on the eye area.” [0099] “1. Skin smoothing... 2. Eye and lip sharpening... 3. Lip color enhancement... 4. Teeth whitening... 5. Eye whitening...”对比文件公开了多种视觉效果,包括皮肤平滑(skin smoothness)、牙齿美白(teeth whitening)、唇色调整(lip color enhancement)、眼睛锐化(eye sharpening)等,这些与特征P中的“皮肤纹理平滑”、“唇膏应用”、“红眼消除”(通过眼睛锐化/美白实现)等效果相对应,因此直接公开。
技术特征Q:所述计算机可读介质和计算机程序代码被配置为在至少一个处理器上运行时至少执行以下操作:通过检测由用户所指向的现实世界中的脸部或者脸部的现实世界中的图像的位置,来检测用户对与脸部相关联的脸部特征的位置指示,使得与脸部相关联的所述脸部特征与现实世界中的脸部或者脸部的现实世界中的图像相关联——不公开[0051] “In some embodiments, user interactivity is provided as an interface allowing user to move fitted points to a desired place.” [0056] “...if the eye corner is located incorrectly by the embodiment, the user can pull the eye corner to the correct location...”基于特征C的论述,对比文件未公开检测用户对现实世界脸部或脸部图像指向的操作,因此特征Q未被公开。
技术特征R:基于所检测到的用户对与脸部相关联的脸部特征的位置指示,提供相对应的计算机所生成的脸部特征的位置的锚定,使得相对应的所述计算机所生成的脸部特征的脸部标记定位能够被锚定在计算机所生成的脸部图像上的相对应位置周围。——不公开[0051] “...user interactivity is provided as an interface allowing user to move fitted points to a desired place. Previously-user-moved points are considered to be trusted and are not moved in the next round of iteration.”基于特征C和D的论述,对比文件未公开特征R所依赖的特定检测方式(即特征C),因此特征R未被直接或隐含公开。
技术特征S:包括:通过检测由用户所指向的现实世界中的脸部或者脸部的现实世界中的图像的位置,来检测用户对与脸部相关联的脸部特征的位置指示,使得与脸部相关联的所述脸部特征与现实世界中的脸部或者脸部的现实世界中的图像相关联——不公开[0051] “In some embodiments, user interactivity is provided as an interface allowing user to move fitted points to a desired place.” [0056] “...if the eye corner is located incorrectly by the embodiment, the user can pull the eye corner to the correct location...”基于特征C的论述,特征S作为方法步骤未被对比文件公开。
技术特征T:基于所检测到的用户对与脸部相关联的脸部特征的位置指示,提供相对应的计算机所生成的脸部特征的位置的锚定,使得相对应的所述计算机所生成的脸部特征的脸部标记定位能够被锚定在计算机所生成的脸部图像上的相对应位置周围。——不公开[0051] “...user interactivity is provided as an interface allowing user to move fitted points to a desired place. Previously-user-moved points are considered to be trusted and are not moved in the next round of iteration.”基于特征C和D的论述,特征T作为方法步骤未被对比文件公开。

<<<A>>><<<B>>><<<E>>><<<F>>><<<H>>><<<I>>><<<J>>><<<N>>><<<O>>><<<P>>>

权利要求与技术特征
Powered by Django

网站备案号:渝ICP备2023012882号


重庆市非显而易见网络科技有限责任公司 A Anti NPE NPE