Abstract: In one embodiment, a computer-implemented method can use a server computer to obtain from a client computer a text input in a query from a user and access in digital data storage coupled to the server computer a plurality of digital images. The computer-implemented method can train a deep learning model to determine a first embedding for the text input and a second embedding of each of the plurality of images. The computer-implemented method can identify one or more relevant images based on the respective similarity of the first embedding to the second embedding. The computer-implemented method can determine image informativeness and confidence scores for information terms of each of the one or more relevant images. The computer-implemented method can transmit to the client computer in response to obtaining the text input, instructions for presenting a user interface comprising the one or more relevant images and the confidence scores.

Abstract: In one embodiment, a computer-implemented method can use a server computer to obtain from a client computer a text input in a query from a user and access in digital data storage coupled to the server computer a plurality of digital images. The computer-implemented method can train a deep learning model to determine a first embedding for the text input and a second embedding of each of the plurality of images. The computer-implemented method can identify one or more relevant images based on a respective similarity of the first embedding to the second embedding. The computer-implemented method can determine image informativeness and confidence scores for information terms of each of the one or more relevant images. The computer-implemented method can transmit to the client computer in response to obtaining the text input, instructions for presenting a user interface comprising the one or more relevant images and the confidence scores.

Abstract: In one embodiment, a computer-implemented method can use a server computer to obtain from a client computer a text input in a query from a user and access in digital data storage coupled to the server computer a plurality of digital images. The computer-implemented method can train a deep learning model to determine a first embedding for the text input and a second embedding of each of the plurality of images. The computer-implemented method can identify one or more relevant images based on a respective similarity of the first embedding to the second embedding. The computer-implemented method can determine image informativeness and confidence scores for information terms of each of the one or more relevant images. The computer-implemented method can transmit to the client computer in response to obtaining the text input, instructions for presenting a user interface comprising the one or more relevant images and the confidence scores.

Abstract: The invention provides an exercise device (10) comprising an elongate member (20) and an engagement member (30) having an engagement region (40), the engagement member defining an aperture (35) therethrough, the elongate member extending through the aperture, such that when the engagement region is under substantially no load, the engagement member is slideable along the elongate member and when the engagement region is placed under load, the engagement member is substantially non-slideable along the elongate member. Also provided is a method of pelvic realignment involving use of said exercise device.

Abstract: The invention provides an exercise device (10) comprising an elongate member (20) and an engagement member (30) having an engagement region (40), the engagement member defining an aperture (35) therethrough, the elongate member extending through the aperture, such that when the engagement region is under substantially no load, the engagement member is slideable along the elongate member and when the engagement region is placed under load, the engagement member is substantially non-slideable along the elongate member. Also provided is a method of pelvic realignment involving use of said exercise device.

Abstract: A properly configured custom action allows a software product to be efficiently and effectively maintained. A custom action is associated with a software product. When the software product is installed on the client device, the custom action facilitates updating on the client. The custom action detects a recache/reinstall command from the source. The custom action identifies the product source, builds a model of the product, and identifies which updates to apply to the software product. When the updates have been identified, the custom action registers the updates on the client computer.

Abstract: A custom action associated with a core installer leverages privileges of the core installer to access satellite installers. The custom action includes code for identifying that a software product includes multiple installers and that the core installer has been cached. The custom action determines the product source. Identified satellite installers are associated with the client device. Pre-generated scripts of the satellite installers are accessed and the scripts are applied to the client to facilitate satellite installer advertisement/assignment. In this manner, the custom action associated with the core installer leverages the privileges of the core installer to install satellite installers on a lockdown client device. The custom action allows multi-installer software products to be installed on a client via a software deployment technology configured for single installer software products. As such, multi-installer software products may be easily deployed and installed on client devices.

Abstract: A refurbishing system for refurbishing a barge includes transport systems that pull the barge out of water and onto a shelf above water level, then pull the barge into a covered blast system and a covered coating system. A reclamation system receives spent abrasive media. The transport systems then move the refurbished barge back into the water. The transport systems move the barge without substantially tilting it or changing its orientation.

Abstract: A first installation package for a first group of files is created for a program. The first group of files include language-neutral files for the program but no language-specific files for the program. A second installation package for a second group of files is also created for the program. The second group of files includes language-specific files for the program.

Abstract: A musician's plectrum having a top and a bottom surface, a first groove on the top surface which runs perpendicular to an axis of symmetry and positioned the user's thumb perpendicular to the axis of symmetry, and a second groove on the bottom surface which runs parallel to the axis of symmetry and positions the user's forefinger parallel to the axis of symmetry, is described. In alternate embodiments, the plectrum can further include a textured surface, a grip support, a beveled striking surface, and/or a wear-resistant coating.