Abstract: Enhanced vectorization of raster images is described. An image vectorization module converts a raster image with bitmapped data to a vector image with vector elements based on mathematical formulas. In some embodiments, spatially-localized control of a vectorization operation is provided to a user. First, the user can adjust an intensity of a denoising operation differently at different areas of the raster image. Second, the user can adjust an automated segmentation by causing one segment to be split into two segments along a zone marked with an indicator tool, such as a brush. Third, the user can adjust an automated segmentation by causing two segments to be merged into a combined segment. The computation of the vector elements is based on the adjusted segmentation. In other embodiments, semantic information gleaned from the raster image is incorporated into the vector image to facilitate manipulation, such as joint selection of multiple vector elements.

Abstract: Maximum curvature techniques are described. In one or more implementations, a curve includes a first data point disposed between second and third data points. The first data point is freely moveable while the second and third data points are constrained from movement.

Abstract: Enhanced vectorization of raster images is described. An image vectorization module converts a raster image with bitmapped data to a vector image with vector elements based on mathematical formulas. In some embodiments, spatially-localized control of a vectorization operation is provided to a user. First, the user can adjust an intensity of a denoising operation differently at different areas of the raster image. Second, the user can adjust an automated segmentation by causing one segment to be split into two segments along a zone marked with an indicator tool, such as a brush. Third, the user can adjust an automated segmentation by causing two segments to be merged into a combined segment. The computation of the vector elements is based on the adjusted segmentation. In other embodiments, semantic information gleaned from the raster image is incorporated into the vector image to facilitate manipulation, such as joint selection of multiple vector elements.

Abstract: Enhanced vectorization of raster images is described. An image vectorization module converts a raster image with bitmapped data to a vector image with vector elements based on mathematical formulas. In some embodiments, spatially-localized control of a vectorization operation is provided to a user. First, the user can adjust an intensity of a denoising operation differently at different areas of the raster image. Second, the user can adjust an automated segmentation by causing one segment to be split into two segments along a zone marked with an indicator tool, such as a brush. Third, the user can adjust an automated segmentation by causing two segments to be merged into a combined segment. The computation of the vector elements is based on the adjusted segmentation. In other embodiments, semantic information gleaned from the raster image is incorporated into the vector image to facilitate manipulation, such as joint selection of multiple vector elements.

Abstract: Enhanced vectorization of raster images is described. An image vectorization module converts a raster image with bitmapped data to a vector image with vector elements based on mathematical formulas. In some embodiments, spatially-localized control of a vectorization operation is provided to a user. First, the user can adjust an intensity of a denoising operation differently at different areas of the raster image. Second, the user can adjust an automated segmentation by causing one segment to be split into two segments along a zone marked with an indicator tool, such as a brush. Third, the user can adjust an automated segmentation by causing two segments to be merged into a combined segment. The computation of the vector elements is based on the adjusted segmentation. In other embodiments, semantic information gleaned from the raster image is incorporated into the vector image to facilitate manipulation, such as joint selection of multiple vector elements.

Abstract: Blending techniques for curve fitting are described. In one or more implementations, an indication is received of three or more data points. A blending factor is computed based on a spatial relationship of the three or more data points to each other. A curve is fit to the three or more data points by blending a plurality or curve fitting techniques using the computed blending factor.

Abstract: Maximum curvature techniques are described. In one or more implementations, a curve includes a first data point disposed between second and third data points. The first data point is freely moveable while the second and third data points are constrained from movement.

Abstract: Parametric curve fitting using maximum curvature techniques are described. In one or more implementations, a parametric curve is fit to a segment of a plurality of data points that includes a first data point disposed between second and third data points by setting a point of maximum curvature for the segment of the curve at the first data point. A result of the fitting is output by the computing device.

Abstract: Image defocus blur estimation techniques are described. In one or more implementations, fixed spatial frequencies that are usable to analyze an image for blur are selected. The selected spatial frequencies are input to a function used to determine frequency responses for pixels of the image. The frequency responses indicate a response of the pixels around a given pixel to the selected spatial frequencies. The spatial frequencies that are selected may be limited to spatial frequencies having a frequency magnitude from a set of discrete values. The discrete values may, for instance, range from a minimum frequency magnitude to a maximum frequency magnitude, and be spaced apart by a frequency magnitude increment. A number of frequencies that are selected at each magnitude may also be based on the frequency magnitude increment.

Abstract: Image defocus blur estimation techniques are described. In one or more implementations, fixed spatial frequencies that are usable to analyze an image for blur are selected. The selected spatial frequencies are input to a function used to determine frequency responses for pixels of the image. The frequency responses indicate a response of the pixels around a given pixel to the selected spatial frequencies. The spatial frequencies that are selected may be limited to spatial frequencies having a frequency magnitude from a set of discrete values. The discrete values may, for instance, range from a minimum frequency magnitude to a maximum frequency magnitude, and be spaced apart by a frequency magnitude increment. A number of frequencies that are selected at each magnitude may also be based on the frequency magnitude increment.

Abstract: Blending techniques for curve fitting are described. In one or more implementations, an indication is received of three or more data points. A blending factor is computed based on a spatial relationship of the three or more data points to each other. A curve is fit to the three or more data points by blending a plurality or curve fitting techniques using the computed blending factor.

Abstract: Parametric curve fitting using maximum curvature techniques are described. In one or more implementations, a parametric curve is fit to a segment of a plurality of data points that includes a first data point disposed between second and third data points by setting a point of maximum curvature for the segment of the curve at the first data point. A result of the fitting is output by the computing device.

Abstract: An image editing application (or a blur classification module thereof) may automatically estimate a coherent defocus blur map from a single input image. The application may represent the blur spectrum as a differentiable function of radius r, and the optimal radius may be estimated by optimizing the likelihood function through a gradient descent algorithm. The application may generate the spectrum function over r through polynomial-based fitting. After fitting, the application may generate look-up tables to store values for the spectrum and for its first and second order derivatives, respectively. The use of these tables in the likelihood optimization process may significantly reduce the computational costs of a given blur estimation exercise. The application may minimize an energy function that includes a data term, a smoothness term, and a smoothness parameter that is adaptive to local image content. The output blur map may be used for image object depth estimation.

Abstract: A blur classification module may compute the probability that a given pixel in a digital image was blurred using a given two-dimensional blur kernel, and may store the computed probability in a blur classification probability matrix that stores probability values for all combinations of image pixels and the blur kernels in a set of likely blur kernels. Computing these probabilities may include computing a frequency power spectrum for windows into the digital image and/or for the likely blur kernels. The blur classification module may generate a coherent mapping between pixels of the digital image and respective blur states, and/or may perform a segmentation of the image into blurry and sharp regions, dependent on values stored in the matrix. Input image data may be pre-processed. Blur classification results may be employed in image editing operations to automatically target image subjects or background regions, or to estimate the depth of image elements.

Abstract: A blur classification module may compute the probability that a given pixel in a digital image was blurred using a given two-dimensional blur kernel, and may store the computed probability in a blur classification probability matrix that stores probability values for all combinations of image pixels and the blur kernels in a set of likely blur kernels. Computing these probabilities may include computing a frequency power spectrum for windows into the digital image and/or for the likely blur kernels. The blur classification module may generate a coherent mapping between pixels of the digital image and respective blur states, and/or may perform a segmentation of the image into blurry and sharp regions, dependent on values stored in the matrix. Input image data may be pre-processed. Blur classification results may be employed in image editing operations to automatically target image subjects or background regions, or to estimate the depth of image elements.

Abstract: A blur classification module may compute the probability that a given pixel in a digital image was blurred using a given two-dimensional blur kernel, and may store the computed probability in a blur classification probability matrix that stores probability values for all combinations of image pixels and the blur kernels in a set of likely blur kernels. Computing these probabilities may include computing a frequency power spectrum for windows into the digital image and/or for the likely blur kernels. The blur classification module may generate a coherent mapping between pixels of the digital image and respective blur states, or may perform a segmentation of the image into blurry and sharp regions, dependent on values stored in the matrix. Input image data may be pre-processed. Blur classification results may be employed in image editing operations to automatically target image subjects or background regions, or to estimate the depth of image elements.

Abstract: A blur classification module may compute the probability that a given pixel in a digital image was blurred using a given two-dimensional blur kernel, and may store the computed probability in a blur classification probability matrix that stores probability values for all combinations of image pixels and the blur kernels in a set of likely blur kernels. Computing these probabilities may include computing a frequency power spectrum for windows into the digital image and/or for the likely blur kernels. The blur classification module may generate a coherent mapping between pixels of the digital image and respective blur states, or may perform a segmentation of the image into blurry and sharp regions, dependent on values stored in the matrix. Input image data may be pre-processed. Blur classification results may be employed in image editing operations to automatically target image subjects or background regions, or to estimate the depth of image elements.

Abstract: Method and apparatus for segmenting a first region and a second region. A method for defining a boundary separating a first region and a second region of a digital image includes determining using a learning machine, based on one or more of the color arrangements, which pixels of the image satisfy criteria for classification as associated with the first region and which pixels of the image satisfy criteria for classification as associated with the second region. The digital image includes one or more color arrangements characteristic of the first region and one or more color arrangements characteristic of the second region. The method includes identifying pixels of the image that are determined not to satisfy the criteria for classification as being associated either with the first region or the second region. The method includes decontaminating the identified pixels to define a boundary between the first and second regions.

Abstract: Methods, systems and apparatus, including computer program products, for processing a computer graphics illustration having pieces of artwork.

Abstract: Methods, systems and apparatus, including computer program products, for processing a computer graphics illustration having pieces of artwork.