公开(公告)号：US20100085372A1

公开(公告)日：2010-04-08

申请号：US11579754

申请日：2005-05-05

申请人： Anat Levin ,  Dani Lischinski ,  Yair Weiss

发明人： Anat Levin ,  Dani Lischinski ,  Yair Weiss

IPC分类号： G09G5/02 ,  G06T7/00

CPC分类号： H04N9/43 ,  G06T11/001 ,  H04N1/465

摘要： A method for colorization of images is presented. The method comprises the following steps: Displaying an image (102); applying multiple color markings (at least two colors are different) to the displayed image (104); automatically coloring the image by applying a constrained optimization to a cost function that is responsive to the marked colors and the intensity differences of the neighborhoods of some of all of the pixels (106); the resultant image is subsequently stored (110), displayed (108), or transferred (112).

摘要翻译： 提出了一种图像着色方法。 该方法包括以下步骤：显示图像（102）; 对所显示的图像（104）应用多个颜色标记（至少两种颜色不同）; 通过将约束优化应用于响应于所有像素（106）中的一些像素的标记颜色和强度差的强度差异的成本函数来自动着色图像; 随后存储（110），显示（108）或传送（112）所得到的图像。

公开(公告)号：US20090278859A1

公开(公告)日：2009-11-12

申请号：US12497800

申请日：2009-07-06

申请人： Yair Weiss ,  Daniel Lischinski ,  Anat Levin

发明人： Yair Weiss ,  Daniel Lischinski ,  Anat Levin

IPC分类号： G09G5/00

CPC分类号： H04N5/272 ,  H04N5/275

摘要： In a method and system for matting a foreground object F having an opacity α constrained by associating a characteristic with selected pixels in an image having a background B, weights are determined for all edges of neighboring pixels for the image and used to build a Laplacian matrix L. The equation α is solved where α=arg min αT Lα s.t.αi=si, ∀iεS, S is the group of selected pixels, and si is the value indicated by the associated characteristic. The equation Ii=αiFi+(1−αi)Bi is solved for F and B with additional smoothness assumptions on F and B; after which the foreground object F may be composited on a selected background B′ that may be the original background B or may be a different background, thus allowing foreground features to be extracted from the original image and copied to a different background.

摘要翻译： 在具有通过将特征与具有背景B的图像中的所选像素相关联而具有不透明度α的前景对象F消除的方法和系统中，为图像的相邻像素的所有边缘确定权重，并用于构建拉普拉斯矩阵 其中α= arg min alphaT Lalpha stalphai = si，∀iepsilonS，S是所选像素的组，si是由相关特征指示的值，求解方程α。 对于F和B求解方程Ii = alphaiFi +（1-alphai）Bi对F和B具有额外的平滑度假设; 之后，可以将前景对象F合成到可以是原始背景B的选定背景B'上，或者可以是不同的背景，从而允许从原始图像提取前景特征并将其复制到不同的背景。

公开(公告)号：US08913074B2

公开(公告)日：2014-12-16

申请号：US11579754

申请日：2005-05-05

申请人： Anat Levin ,  Dani Lischinski ,  Yair Weiss

发明人： Anat Levin ,  Dani Lischinski ,  Yair Weiss

IPC分类号： G09G5/02 ,  G06T11/00 ,  H04N9/43

CPC分类号： H04N9/43 ,  G06T11/001 ,  H04N1/465

摘要： A method for colorization of images is presented. The method comprises the following steps: Displaying an image (102); applying multiple color markings (at least two colors are different) to the displayed image (104); automatically coloring the image by applying a constrained optimization to a cost function that is responsive to the marked colors and the intensity differences of the neighborhoods of some of all of the pixels (106); the resultant image is subsequently stored (110), displayed (108), or transferred (112).

摘要翻译： 提出了一种图像着色方法。 该方法包括以下步骤：显示图像（102）; 对所显示的图像（104）应用多个颜色标记（至少两种颜色不同）; 通过将约束优化应用于响应于所有像素（106）中的一些像素的标记颜色和强度差的强度差异的成本函数来自动着色图像; 随后存储（110），显示（108）或传送（112）所得到的图像。

公开(公告)号：US07692664B2

公开(公告)日：2010-04-06

申请号：US12497800

申请日：2009-07-06

申请人： Yair Weiss ,  Daniel Lischinski ,  Anat Levin

发明人： Yair Weiss ,  Daniel Lischinski ,  Anat Levin

IPC分类号： G09G5/02

CPC分类号： H04N5/272 ,  H04N5/275

摘要： In a method and system for matting a foreground object F having an opacity α constrained by associating a characteristic with selected pixels in an image having a background B, weights are determined for all edges of neighboring pixels for the image and used to build a Laplacian matrix L. The equation α is solved where α=arg min αT Lα s.t.αi=si, ∀iεS, S is the group of selected pixels, and si is the value indicated by the associated characteristic. The equation Ii=αiFi+(1−αi)Bi is solved for F and B with additional smoothness assumptions on F and B; after which the foreground object F may be composited on a selected background B′ that may be the original background B or may be a different background, thus allowing foreground features to be extracted from the original image and copied to a different background.

摘要翻译： 在通过将特征与具有背景B的图像中的所选像素相关联而具有不透明度α的前景对象F的方法和系统中，为图像的相邻像素的所有边缘确定权重，并用于构建拉普拉斯矩阵 求解方程α其中α= arg minαTLαstαi= si，∀i＆egr; S，S是所选像素的组，si是由相关特征表示的值。 对F和B求解方程Ii =αiFi+（1-αi）Bi，对F和B进行额外的平滑假设; 之后，可以将前景对象F合成到可以是原始背景B的选定背景B'上，或者可以是不同的背景，从而允许从原始图像提取前景特征并将其复制到不同的背景。

公开(公告)号：US20070156471A1

公开(公告)日：2007-07-05

申请号：US11289343

申请日：2005-11-29

申请人： Baback Moghaddam ,  Yair Weiss ,  Shmuel Avidan

发明人： Baback Moghaddam ,  Yair Weiss ,  Shmuel Avidan

IPC分类号： G06F9/44 ,  G06F17/50 ,  G06Q40/00

CPC分类号： G06K9/6234 ,  G06Q40/00 ,  G06Q40/04

摘要： A method maximizes a candidate solution to a cardinality-constrained combinatorial optimization problem of sparse principal component analysis. An approximate method has as input a covariance matrix A, a candidate solution, and a sparsity parameter k. A variational renormalization for the candidate solution vector x with regards to the eigenvalue structure of the covariance matrix A and the sparsity parameter k is then performed by means of a sub-matrix eigenvalue decomposition of A to obtain a variance maximized k-sparse eigenvector x that is the best possible solution. Another method solves the problem by means of a nested greedy search technique that includes a forward and backward pass. An exact solution to the problem initializes a branch-and-bound search with an output of a greedy solution.

摘要翻译： 一种方法将候选解最大化为稀疏主分量分析的基数约束组合优化问题。 近似方法具有协方差矩阵A，候选解和稀疏参数k作为输入。 然后通过A的子矩阵特征值分解来执行关于协方差矩阵A的特征值结构和稀疏参数k的候选解矢量x的变分重归一化，以获得方差最大化的k-稀疏特征向量x，其中 是最好的解决方案。 另一种方法通过嵌套的贪婪搜索技术来解决问题，该技术包括前进和后退。 问题的确切解决方案使用贪心解决方案的输出初始化分支搜索。

公开(公告)号：US20070165966A1

公开(公告)日：2007-07-19

申请号：US11487482

申请日：2006-07-17

申请人： Yair Weiss ,  Daniel Lischinski ,  Anat Levin

发明人： Yair Weiss ,  Daniel Lischinski ,  Anat Levin

IPC分类号： G06K9/36 ,  G06K9/34

CPC分类号： H04N5/272 ,  H04N5/275

摘要： In a method and system for matting a foreground object F having an opacity α constrained by associating a characteristic with selected pixels in an image having a background B, weights are determined for all edges of neighboring pixels for the image and used to build a Laplacian matrix L. The equation α is solved where α=arg min αT Lα s.t.αi=si, ∀i ∈ S, S is the group of selected pixels, and si is the value indicated by the associated characteristic. The equation Ii=αiFi+(1−αi)Bi is solved for F and B with additional smoothness assumptions on F and B; after which the foreground object F may be composited on a selected background B′ that may be the original background B or may be a different background, thus allowing foreground features to be extracted from the original image and copied to a different background.

摘要翻译： 在具有通过将特征与具有背景B的图像中的所选像素相关联而具有不透明度α的前景对象F消除的方法和系统中，为图像的相邻像素的所有边缘确定权重，并且用于构建拉普拉斯矩阵 方程式α被求解，其中α= arg minαL pha pha∈∈∈∈∈∈S S S is is is is is is is is 所选择的像素组以及相关联的特征所指示的值。 方程式I> i i i）））<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< 对于F和B，对F和B进行额外的平滑假设求解/ SUB> 之后，可以将前景对象F合成到可以是原始背景B的选定背景B'上，或者可以是不同的背景，从而允许从原始图像提取前景特征并将其复制到不同的背景。

公开(公告)号：US20070122041A1

公开(公告)日：2007-05-31

申请号：US11440825

申请日：2006-05-25

申请人： Baback Moghaddam ,  Yair Weiss ,  Shmuel Avidan

发明人： Baback Moghaddam ,  Yair Weiss ,  Shmuel Avidan

IPC分类号： G06K9/62 ,  G06K9/52 ,  G06K9/46

CPC分类号： G06K9/6234

摘要： A computer implemented method maximizes candidate solutions to a cardinality-constrained combinatorial optimization problem of sparse linear discriminant analysis. A candidate sparse solution vector x with k non-zero elements is inputted, along with a pair of covariance matrices A, B measuring between-class and within-class covariance of binary input data to be classified, the sparsity parameter k denoting a desired cardinality of a final solution vector. A variational renormalization of the candidate solution vector x is performed with regards to the pair of covariance matrices A, B and the sparsity parameter k to obtain a variance maximized discriminant eigenvector {circumflex over (x)} with cardinality k that is locally optimal for the sparsity parameter k and zero-pattern of the candidate sparse solution vector x, and is the final solution vector for the sparse linear discriminant analysis optimization problem. Another method solves the initial problem of finding a candidate sparse solution by means of a nested greedy search technique that includes a forward and backward pass. Another method, finds an exact and optimal solution to the general combinatorial problem by first finding a candidate by means of the previous nested greedy search technique and then using this candidate to initialize a branch-and-bound algorithm which gives the optimal solution.

摘要翻译： 计算机实现的方法使候选解决方案最大化为稀疏线性判别分析的基数约束组合优化问题。 输入具有k个非零元素的候选稀疏解矢量x，以及测量要分类的二进制输入数据的类间和类内协方差之间的一对协方差矩阵A，B，稀疏参数k表示期望的基数 的最终解矢量。 对于一对协方差矩阵A，B和稀疏参数k来执行候选解矢量x的变分重正化，以获得对于稀疏参数k和零稀疏参数k是局部最优的基数k的方差最大化判别特征向量x， 候选稀疏解向量x的模式，是稀疏线性判别分析优化问题的最终解向量。 另一种方法解决了通过包含向前和向后遍的嵌套贪婪搜索技术找到候选稀疏解的初始问题。 另一种方法，通过首先通过先前的嵌套贪婪搜索技术找到候选者，然后使用该候选来初始化给出最优解的分支绑定算法，找到一般组合问题的精确和最优解。

公开(公告)号：US06745157B1

公开(公告)日：2004-06-01

申请号：US09586281

申请日：2000-06-02

申请人： Yair Weiss ,  William T. Freeman ,  Jonathan S. Yedidia

发明人： Yair Weiss ,  William T. Freeman ,  Jonathan S. Yedidia

IPC分类号： G06F1710

CPC分类号： G06F17/10

摘要： A method determines the probabilities of states of a system represented by a model including of nodes connected by links. Each node represents possible states of a corresponding part of the system, and each link represents statistical dependencies between possible states of related nodes. The nodes are grouped into arbitrary sized clusters such that every node is included in at least one cluster. A minimal number of marginalization constraints to be satisfied between the clusters are determined. A super-node network is constructed so that each cluster of nodes is represented by exactly one super-node. Super-nodes that share one of the marginalization constraints are connected by super-links. The super-node network is searched to locate closed loops of super-nodes containing at least one common node. A normalization operator for each closed loop is determined, and messages between the super-nodes are defined. Initial values are assigned to the messages, and the messages between super-nodes are updated using standard belief propagation. The messages are replaced by associated normalized values using the corresponding normalization operator, and approximate probabilities of the states of the system are determined from the messages when a termination condition is reached.

摘要翻译： 一种方法确定由包括通过链接连接的节点的模型所表示的系统的状态的概率。 每个节点表示系统的对应部分的可能状态，并且每个链路表示相关节点的可能状态之间的统计依赖性。 节点被分组成任意大小的集群，使得每个节点被包括在至少一个集群中。 确定要在群集之间满足的最小数量的边缘化约束。 构建超节点网络，使得每个节点簇由正好一个超节点表示。 分享边缘化约束的超级节点通过超级链接连接。 搜索超节点网络以定位包含至少一个公共节点的超节点的闭环。 确定每个闭环的归一化运算符，并定义超节点之间的消息。 初始值分配给消息，超级节点之间的消息使用标准置信传播进行更新。 使用相应的归一化算子将消息替换为关联的归一化值，并且当达到终止条件时，从消息确定系统的状态的近似概率。