Certain aspects relate to systems and techniques for image metadata enrichment. Image metadata enrichment can be accomplished by taking a variety of structured and unstructured data associated with an image and integrating some or all of the structured and unstructured data into the metadata of the image. The enriched image metadata renders the image capable of being placed against other leveraging technologies, for example searching and feature-based sorting.

An information processing system includes a server apparatus that processes a document, and a client apparatus that provides a user with a result of processing the document, in which processing of the document is performed in the server apparatus and the client apparatus in response to an instruction from the user, in a case where the processing of the document in the client apparatus is completed faster than the processing of the document in the server apparatus, a result of the processing of the document is transmitted from the client apparatus to the server apparatus, and in a case where an instruction for the document is given by another client apparatus different from the client apparatus, the result of the processing of the document is transmitted from the server apparatus to the other client apparatus.

Techniques are disclosed for improving transfer speed for a plurality of files (e.g., image files) by using a Markov decision process to determine an optimal number of parallel instances of transfer stages and optimal file batch sizes for each instance. The transfer (e.g., import or export) operation involves different stages that are each optimized using the algorithm. The stages include a file fetch operation, a file processing operation, and a database update operation. Each of the stages may have multiple parallel instances to process many files at the same time. The Markov decision process uses a reward structure to determine the optimal number of parallel instances for each stage and the number of files operated on at each instance at any given moment in time. The process is dynamic and adaptable to any system environment since it does not rely on any particular hardware or operating system configuration.

Techniques are disclosed for improving transfer speed for a plurality of files (e.g., image files) by using a Markov decision process to determine an optimal number of parallel instances of transfer stages and optimal file batch sizes for each instance. The transfer (e.g., import or export) operation involves different stages that are each optimized using the algorithm. The stages include a file fetch operation, a file processing operation, and a database update operation. Each of the stages may have multiple parallel instances to process many files at the same time. The Markov decision process uses a reward structure to determine the optimal number of parallel instances for each stage and the number of files operated on at each instance at any given moment in time. The process is dynamic and adaptable to any system environment since it does not rely on any particular hardware or operating system configuration.

The methods and systems are designed to utilize an integrated combination of just in time, just in place, and just on device actions connected to an image recognition process for monitoring criminals who are probation, offenders who are on parole, sex offenders, and witnesses under protection by law enforcement.

Modeling semantic concepts in an embedding space as distributions is described. In the embedding space, both images and text labels are represented. The text labels describe semantic concepts that are exhibited in image content. In the embedding space, the semantic concepts described by the text labels are modeled as distributions. By using distributions, each semantic concept is modeled as a continuous cluster which can overlap other clusters that model other semantic concepts. For example, a distribution for the semantic concept “apple” can overlap distributions for the semantic concepts “fruit” and “tree” since can refer to both a fruit and a tree. In contrast to using distributions, conventionally configured visual-semantic embedding spaces represent a semantic concept as a single point. Thus, unlike these conventionally configured embedding spaces, the embedding spaces described herein are generated to model semantic concepts as distributions, such as Gaussian distributions, Gaussian mixtures, and so on.

A search engine is modified to perform increasingly precise image searching using iterative Natural Language (NL) interactions. From an NL search input, the modification extracts a set of input features, which includes a set of response features corresponding to an NL statement in the NL search input and a set of image features from a seed image in the NL search input. The modification performs image analysis on an image result in a result set of a query including at least some of the input features. In a next iteration of NL interactions, at least some of the result set is provided. An NL response in the iteration is added to a cumulative NL basis, and a revised result set is provided, which includes a new image result corresponding to a new response feature extracted from the cumulative NL basis.

A training data acquisition method and device, a server and a storage medium are provided. The training data acquisition method is applied to a classifier and includes the following steps: obtaining an image search target according to an input of a user; providing images to the user according to the image search target, to display the images; and selecting at least one image from the displayed images, and determining a target-classification pair as training data according to the at least one image; where the target-classification pair includes the image search target and an entity-based classification of the at least one image.

A training data acquisition method and device, a server and a storage medium are provided. The training data acquisition method is applied to a classifier and includes the following steps: obtaining an image search target according to an input of a user; providing images to the user according to the image search target, to display the images; and selecting at least one image from the displayed images, and determining a target-classification pair as training data according to the at least one image; where the target-classification pair includes the image search target and an entity-based classification of the at least one image.

An embodiment of the invention is directed to a system that comprises one or more processors and a memory communicatively coupled to the one or more processors. Herein, the memory including logic that, upon executing by the one or more processors and in response to receiving information that uniquely identifies a vehicle, automatically generating a plurality of images of the vehicle. A vehicle identification number (VIN) may be used as the information that uniquely identifies a vehicle and causes automatic generation of the interior and exterior images for that specific vehicle.