Abstract: Methods, apparatus, and computer-readable storage media for k-NN re-ranking. Based on retrieved images and localized objects, a k-NN re-ranking method may use the k-nearest neighbors of a query to refine query results. Given the top k retrieved images and their localized objects, each k-NN object may be used as a query to perform a search. A database image may have different ranks when using those k-nearest neighbors as queries. Accordingly, a new score for each database image may be collaboratively determined by those ranks, and re-ranking may be performed using the new scores to improve the search results. The k-NN re-ranking technique may be performed two or more times, each time on a new set of k-nearest neighbors, to further refine the search results.

Abstract: Each of multiple images is analyzed to determine how the colors of the pixels of the image are distributed throughout the color space of the image. Different covariance based characteristics of the image are determined that identify a direction, as well as magnitude in each direction, of the distribution of colors of the image pixels. These different covariance based characteristics that are determined for an image can be saved as associated with the image, allowing the characteristics to be accessed and used as a basis for searching the images to identify particular types of images. These different covariance based characteristics can also be used to order the images identified by a search.

Abstract: Query object localization, segmentation, and retrieval are disclosed. A query image may be received that includes a query object. Based on respective spatially constrained similarity measures between the query image and a plurality of images from an image database, at least some of the plurality of images may be identified and/or retrieved and a location of the query object in the query image may be estimated. The query object may then be automatically segmented from the query image based on the estimated query object location. In some embodiments, the retrieval, localization and/or segmentation may be iterated.

Abstract: Methods, apparatus, and computer-readable storage media for object retrieval and localization that employ a spatially-constrained similarity model. A spatially-constrained similarity measure may be evaluated by a voting-based scoring technique. Object retrieval and localization may thus be achieved without post-processing. The spatially-constrained similarity measure may handle object rotation, scaling and view point change. The similarity measure can be efficiently calculated by the voting-based method and integrated with inverted files. The voting-based scoring technique may simultaneously retrieve and localize a query object in a collection of images such as an image database. The object retrieval and localization technique may, for example, be implemented with a k-nearest neighbor (k-NN) re-ranking method in or as a retrieval method, system or module. The k-NN re-ranking method may be applied to improve query results of the object retrieval and localization technique.

Abstract: Various embodiments of methods and apparatus for facial retouching are disclosed. In one embodiment, a face in an input image is detected. One or more transformation parameters for the detected face are estimated based on a profile model. The profile model is applied to obtain a set of feature points for each facial component of the detected face. Global and component-based shape models are applied to generate feature point locations of each facial component of the detected face.

Abstract: Each of multiple images is analyzed to determine how the colors of the pixels of the image are distributed throughout the color space of the image. Different covariance based characteristics of the image are determined that identify a direction, as well as magnitude in each direction, of the distribution of colors of the image pixels. These different covariance based characteristics that are determined for an image can be saved as associated with the image, allowing the characteristics to be accessed and used as a basis for searching the images to identify particular types of images. These different covariance based characteristics can also be used to order the images identified by a search.

Abstract: Various embodiments of methods and apparatus for facial retouching are disclosed. In one embodiment, a face in an input image is detected. Independent sets of feature points are detected for respective facial feature components. A plurality of masks for each of the facial feature components is generated. Using the plurality of masks, retouch effects are performed to the facial feature components. Some embodiments provide for user interaction to constrain the mask generation.

Abstract: For each image of a set of images, the each image is characterized with a set of fixed-orientation texture descriptors and a set of color descriptors. The set of images is indexed in a color index and a texture index. Similarly, a query image is characterized with a set of fixed-orientation texture descriptors. The set of fixed orientation texture descriptors of the query image includes a set of fixed orientation descriptors for each of a set of rotated query images, and a set of color descriptors of the query image. A rotated local Bag-of-Features (BoF) operation is performed upon the set of rotated query images and the set of images. Each of the set of images is ranked based on the rotated local Bag-of-Features operation.

Abstract: Each image of a set of images is characterized with a set of sparse feature descriptors and a set of dense feature descriptors. In some embodiments, both the set of sparse feature descriptors and the set of dense feature descriptors are calculated based on a fixed rotation for computing texture descriptors, while color descriptors are rotation invariant. In some embodiments, the descriptors of both sparse and dense features are then quantized into visual words. Each database image is represented by a feature index including the visual words computed from both sparse and dense features. A query image is characterized with the visual words computed from both sparse and dense features of the query image. A rotated local Bag-of-Features (BoF) operation is performed upon a set of rotated query images against the set of database images. Each of the set of images is ranked based on the rotated local Bag-of-Features operation.

Abstract: Various embodiments of methods and apparatus for feature point localization are disclosed. An object in an input image may be detected. A profile model may be applied to determine feature point locations for each object component of the detected object. Applying the profile model may include globally optimizing the feature points for each object component to find a global energy minimum. A component-based shape model may be applied to update the respective feature point locations for each object component.

Abstract: Various embodiments of methods and apparatus for feature point localization are disclosed. A profile model and a shape model may be applied to an object in an image to determine locations of feature points for each object component. Input may be received to move one of the feature points to a fixed location. Other ones of the feature points may be automatically adjusted to different locations based on the moved feature point.

Abstract: Query object localization, segmentation, and retrieval are disclosed. A query image may be received that includes a query object. Based on respective spatially constrained similarity measures between the query image and a plurality of images from an image database, at least some of the plurality of images may be identified and/or retrieved and a location of the query object in the query image may be estimated. The query object may then be automatically segmented from the query image based on the estimated query object location. In some embodiments, the retrieval, localization and/or segmentation may be iterated.

Abstract: Methods and systems for performing fast, large-scale, localized Bag-of-Features (Local BoF) retrieval are disclosed. In some embodiments, a method may include receiving a query image and ranking each image of a large set of database images as a function of its similarity to the query image with a Local BoF operation. A Local BoF operation may be configured to localize, for each ranked image, a region that has a highest similarity to the query image. As such, the systems and methods described herein may be suitable for use in large-scale image search and retrieval or categorization operations that may identify objects of interest with arbitrary rotations, significantly different viewpoints, in the presence of clutter. In some embodiments, systems and methods described herein may be used as building blocks of various computer vision and image processing applications including, for example, object recognition and categorization, 3D modeling, mapping, navigation, gesture interfaces, etc.

Abstract: For each image of a set of images, the each image is characterized with a set of fixed-orientation texture descriptors and a set of color descriptors. The set of images is indexed in a color index and a texture index. Similarly, a query image is characterized with a set of fixed-orientation texture descriptors. The set of fixed orientation texture descriptors of the query image includes a set of fixed orientation descriptors for each of a set of rotated query images, and a set of color descriptors of the query image. A rotated local Bag-of-Features (BoF) operation is performed upon the set of rotated query images and the set of images. Each of the set of images is ranked based on the rotated local Bag-of-Features operation.

Abstract: Methods and systems for fast, large scale, high-dimensional searches are described. In some embodiments, a method comprises transforming components of a high-dimensional image descriptor into transformed components in a transform domain, allocating one or more bits available within a bit budget to a given transformed component within a first subset of transformed components as a function of a variance of the given transformed component, independently quantizing each transformed component within the first subset of transformed components, generating a compact representation of the high-dimensional image descriptor based, at least in part, on the independently quantized components, and evaluating a nearest neighbor search operation based, at least in part, on the compact representation of the high-dimensional image descriptor.

Abstract: Each image of a set of images is characterized with a set of sparse feature descriptors and a set of dense feature descriptors. In some embodiments, both the set of sparse feature descriptors and the set of dense feature descriptors are calculated based on a fixed rotation for computing texture descriptors, while color descriptors are rotation invariant. In some embodiments, the descriptors of both sparse and dense features are then quantized into visual words. Each database image is represented by a feature index including the visual words computed from both sparse and dense features. A query image is characterized with the visual words computed from both sparse and dense features of the query image. A rotated local Bag-of-Features (BoF) operation is performed upon a set of rotated query images against the set of database images. Each of the set of images is ranked based on the rotated local Bag-of-Features operation.

Abstract: Methods and systems for fast, large scale, high-dimensional searches are described. In some embodiments, a method comprises transforming components of a high-dimensional image descriptor into transformed components in a transform domain, allocating one or more bits available within a bit budget to a given transformed component within a first subset of transformed components as a function of a variance of the given transformed component, independently quantizing each transformed component within the first subset of transformed components, generating a compact representation of the high-dimensional image descriptor based, at least in part, on the independently quantized components, and evaluating a nearest neighbor search operation based, at least in part, on the compact representation of the high-dimensional image descriptor.

Abstract: Embodiments of methods and systems for stereo-aware image editing are described. A three-dimensional model of a stereo scene is built from one or more input images. Camera parameters for the input images are computed. The three-dimensional model is modified. In some embodiments, the modifying the three-dimensional model includes modifying one or more of the images and applying results of the modifying one or more of the images to corresponding model vertices. The scene is re-rendered from the camera parameters to produce an edited stereo pair that is consistent with the three-dimensional model.

Abstract: Methods, apparatus, and computer-readable storage media for object retrieval and localization that employ a spatially-constrained similarity model. A spatially-constrained similarity measure may be evaluated by a voting-based scoring technique. Object retrieval and localization may thus be achieved without post-processing. The spatially-constrained similarity measure may handle object rotation, scaling and view point change. The similarity measure can be efficiently calculated by the voting-based method and integrated with inverted files. The voting-based scoring technique may simultaneously retrieve and localize a query object in a collection of images such as an image database. The object retrieval and localization technique may, for example, be implemented with a k-nearest neighbor (k-NN) re-ranking method in or as a retrieval method, system or module. The k-NN re-ranking method may be applied to improve query results of the object retrieval and localization technique.

Abstract: Methods, apparatus, and computer-readable storage media for k-NN re-ranking. Based on retrieved images and localized objects, a k-NN re-ranking method may use the k-nearest neighbors of a query to refine query results. Given the top k retrieved images and their localized objects, each k-NN object may be used as a query to perform a search. A database image may have different ranks when using those k-nearest neighbors as queries. Accordingly, a new score for each database image may be collaboratively determined by those ranks, and re-ranking may be performed using the new scores to improve the search results. The k-NN re-ranking technique may be performed two or more times, each time on a new set of k-nearest neighbors, to further refine the search results.