Provided are a device and operating method thereof for obtaining compression ratio information for recognizing a target object in an image using a deep neural network model, and compressing an image using the compression ratio information and encoding the compressed image. According to an embodiment of the present disclosure, there is provided a device that receives an image via at least one camera or a communication interface, obtains a feature map for detecting a target object in the received image, outputs a compression ratio for correctly recognizing the target object in the image by inputting the image and the feature map to a deep neural network model composed of pre-trained model parameters, and generates a bitstream by compressing the image using the output compression ratio and encoding the compressed image.

Apparatus and methods for identification of a coded pattern visible to a computerized imaging apparatus while invisible or inconspicuous to human eyes. A pattern and/or marking may serve to indicate identity of an object, and/or the relative position of the pattern to a viewer. While some solutions exist for identifying patterns (for example, QR codes), they may be visually obtrusive to a human observer due to visual clutter. In exemplary implementations, apparatus and methods are capable of generating patterns with sufficient structure to be used for either discrimination or some aspect of localization, while incorporating spectral properties that are more aesthetically acceptable such as being: a) imperceptible or subtle to the human observer and/or b) aligned to an existing acceptable visual form, such as a logo. In one variant, a viewer comprises an imaging system comprised as a processor and laser scanner, or camera, or moving photodiode.

In a method of optical proximity correction (OPC) modeling, a resist image (RI) model is generated from an aerial image (AI) of a pattern. A light intensity of a portion having a level lower than a truncation level is replaced with the truncation level in an image profile of the RI model. The image profile is smoothed to remove a sharp point in the image profile. A Laplacian kernel is applied to the image profile to generate a contour image profile. A portion of the contour image profile having a value lower than a given level is truncated. A radius of curvature kernel is applied to the contour image profile. A reciprocal number of the radius of curvature is applied to the RI model.

A method and apparatus for performing gesture recognition. In one embodiment of the invention, the method includes the steps of receiving one or more raw frames from one or more cameras, each of the one or more raw frames representing a time sequence of images, determining one or more regions of the one or more received raw frames that comprise highly textured regions, segmenting the one or more determined highly textured regions in accordance textured features thereof to determine one or more segments thereof, determining one or more regions of the one or more received raw frames that comprise other than highly textured regions, and segmenting the one or more determined other than highly textured regions in accordance with color thereof to determine one or more segments thereof. One or more of the segments are then tracked through the one or more raw frames representing he time sequence of images.

Approaches are described for shape-based graphics search. Each graphics object of a set of graphics objects is analyzed. The analyzing includes determining an outline of the graphics object from graphics data that forms the graphics object. The outline of the graphics object is sampled resulting in sampled points that capture the outline of the graphics object. A shape descriptor of the graphics object is determined which captures local and global geometric properties of the sampled points. Search results of a search query are determined based on a comparison between a shape descriptor of a user identified graphics object and the shape descriptor of at least one graphics object of the set of graphics objects. At least one of the search results can be presented on a user device associated with the search query.

Methods, systems, and computer programs are presented for adding new features to a network service. A method includes receiving an image depicting an object of interest. A category set is determined for the object of interest and an image signature is generated for the image. Using the category set and the image signature, the method identifies a set of publications within a publication database and assigns a rank to each publication. The method causes presentation of the ranked list of publications at a computing device from which the image was received.

The present disclosure relates to the technical field of image processing, and in particular to an image description generation method, apparatus and system, and a medium and an electronic device. The method comprises: acquiring one or more image region features in a target image, and obtaining a current input vector by performing a mean pooling on the image region features; obtaining respective outer product vectors of the image region features by respectively linearly fusing the current input vector and each of the image region features; calculating, based on the respective outer product vectors of the image region features, an attention distribution of the image region features in a spatial dimension and an attention distribution of the image region features in a channel dimension; and generating an image description of the target image based on the attention distribution of the image region features in the spatial dimension and the attention distribution of the image region features in the channel dimension.

A vertebra localization and identification method includes: receiving one or more images of vertebrae of a spine; applying a machine learning model on the one or more images to generate three-dimensional (3-D) vertebra activation maps of detected vertebra centers; performing a spine rectification process on the 3-D vertebra activation maps to convert each 3-D vertebra activation map into a corresponding one-dimensional (1-D) vertebra activation signal; performing an anatomically-constrained optimization process on each 1-D vertebra activation signal to localize and identify each vertebra center in the one or more images; and outputting the one or more images, wherein on each of the one or more outputted images, a location and an identification of each vertebra center are specified.

An information processing system includes an imaging unit that generates an image signal by imaging and an information processing device. The information processing device performs at least any one of plural kinds of image processing on a taken image corresponding to the image signal. The information processing device specifies an object corresponding to a partial image included in the taken image on the basis of a state of the object corresponding to the partial image included in the taken image or a degree of reliability given to a processing result of the performed image processing.

A 3D shape matching method and a 3D shape matching device based on 3D local feature description using SGHs are provided. In the method, the spherical neighborhood of the feature point is not only divided based on space but also divided based on geometry, the spherical neighborhood of the feature point is not only divided based on the radial direction and the azimuth respectively but also divided based on the elevation, and the spherical neighborhood of the feature point is not only divided based on the deviation angle deviating from the z axis but also divided based on the deviation angle deviating from the x axis. When the deviation angle deviating from the z axis of the spherical neighborhood is divided, the deviation angle is divided more densely where it is closer to the positive direction of the z axis.