Abstract: Methods, apparatus, electronic devices, and computer-readable storage media for calibrating image acquiring devices are provided. In one aspect, a computer-implemented method includes: obtaining a scene image of a preset scene acquired by an image acquiring device disposed on a traveling device, the preset scene including at least two parallel lines, the traveling device being located between adjacent two parallel lines, and sides of the traveling device being substantially parallel to the two parallel lines; based on the scene image, determining two line segments, and first coordinate information of a plurality of target reference points on each line segment of the two line segments in a pixel coordinate system and second coordinate information of the plurality of target reference points in a world coordinate system; and determining a homography matrix corresponding to the image acquiring device based on the first coordinate information and the second coordinate information.

Abstract: One example of a method of indexing a plurality of images includes, for each of the plurality of images, generating a feature vector for the image, applying a trained set of classifiers to the feature vector to generate a score vector for the image, and, based on the score vector and a set of category word vectors, producing a variable number of semantic embedding vectors for the image.

Abstract: Methods, devices related to image retrieval are described herein. A method for performing an image search includes receiving a textual search term from a user, determining a first semantic representation of the textual search term, and determining differences between the first semantic representation and multiple semantic representations that correspond to a plurality of images. Each of the multiple semantic representations is determined based on combining local and global information of a corresponding image. The local information indicates a correlation between features of the corresponding image, and the global information indicates a correspondence between the features of the corresponding image and one or more semantic categories. The method also includes retrieving one or more images as search results in response to the textual search term based on the determined differences.

Abstract: An image reconstruction method includes: an image feature corresponding to a first image in video data and an image feature corresponding to each second image adjacent to the first image are acquired; feature optimization processing is performed on the image feature of the first image and the image feature of the second image to obtain a first optimized feature corresponding to the first image and a second optimized feature corresponding to the second image; feature fusion processing is performed on the first optimized feature and the second optimized feature according to an adjacency matrix between the first optimized feature and the second optimized feature to obtain a fused feature; and image reconstruction processing is performed on the first image using the fused feature to obtain a reconstructed image corresponding to the first image.

Abstract: A system for determining an optimal forecasting hierarchy includes setting a plurality of hierarchies by setting each hierarchy in terms of a hierarchy structure. The system also includes generating top level forecasts for a training period and a testing period, each top level forecast being based upon an aggregation of the base forecast levels. The system further includes calculating a forecast error. An optimization routine includes determining an optimal hierarchy and an optimal base level, both being based upon a smallest associated forecast error. The optimization routine also includes calculating a deficiency ratio for each level within each hierarchy by comparing forecast errors associated with the optimal base level and the optimal hierarchy to forecast errors associated with each of the other base levels and hierarchies. The optimal forecasting hierarchy is determined by comparing the deficiency ratio to a significance level.

Abstract: Selecting one or more entities from a plurality of candidate entities subject to one or more constraints is accomplished by establishing upper and lower profit-margin boundary values and determining an optimal profit-margin value within the boundary values using an iterative process. The iterative process includes one or more iterations of: setting a current profit-margin candidate based on a current upper profit-margin boundary value and a current lower profit-margin boundary value, determining whether the current profit-margin candidate value satisfies the constraint(s), adjusting the current upper or lower profit-margin boundary value based on the current profit-margin candidate value, and determining whether to use the current profit-margin candidate value as the optimal profit-margin value based the current upper profit-margin boundary value and the current lower profit-margin boundary value.

Abstract: In one embodiment of the present invention, a method for forecasting a Last Time Buy quantity for a service part using a low-pass filter approach is provided. The method includes accessing input data including: (1) a service lifespan beginning at the end of mass production of a product associated with the service part; (2) data reflecting past accumulated production of the associated product for at least a first period beginning at the end of mass production of the associated product; and (3) an order data series reflecting past quantities of the service part ordered in each of a succession of periods beginning with a second period that immediately follows the first period and ending with a current period. A low-pass filter is applied to at least a portion of the order data series to extract low frequency components representing a smoothed order data series. A local maximum is sought within the smoothed order data series subject to predetermined conditions.

Abstract: Within each iteration of an iterative process: (1) a low-pass filter is applied to an actual sales volumes series to extract low frequency components representing a base sales volume series for the iteration; and (2) a locally optimal base sales volume series is determined. A globally optimal base sales volume series is selected from among the locally optimal base sales volume series, comprising an estimated base sales volume for each time period. One or more of the estimated base sales volumes is made available for use in connection with at least one business analysis.

Abstract: Within each iteration of an iterative process: (1) a low-pass filter is applied to an actual sales volumes series to extract low frequency components representing a base sales volume series for the iteration; and (2) a locally optimal base sales volume series is determined. A globally optimal base sales volume series is selected from among the locally optimal base sales volume series, comprising an estimated base sales volume for each time period. One or more of the estimated base sales volumes is made available for use in connection with at least one business analysis.

Abstract: Within each iteration of an iterative process: (1) a low-pass filter is applied to an actual sales volumes series to extract low frequency components representing a base sales volume series for the iteration; and (2) a locally optimal base sales volume series is determined. A globally optimal base sales volume series is selected from among the locally optimal base sales volume series, comprising an estimated base sales volume for each time period. One or more of the estimated base sales volumes is made available for use in connection with at least one business analysis.