Abstract: The cryptographic system allows a cloud service provider to persist customer data on behalf of an intermediary service provider, but provide end-to-end encryption such that cloud service provider is not able to read the data. A Secrets Vault provides a cryptographically enforced mechanism by which Secrets are protected and only accessible by authorised users or services, and access to Secrets is cryptographically provable. An entity with a valid Credential has an associated Key Pair. The Credential is used to encrypt/wrap the Key Pair (namely the private key of the Key Pair, as by definition public Keys are designed to be publicly accessible). This allows the Key Pair to be stored online, in an encrypted form. Secrets have corresponding Secret Keys which are used to symmetrically encrypt the Secrets. The Secret Keys are then asymmetrically encrypted using public-private key-pairs.

Abstract: Image annotation includes: accessing initial object prediction information associated with an image, wherein the initial object prediction information includes a plurality of initial predictions associated with a plurality of objects in the image, including bounding box information associated with the plurality of objects; presenting the image and at least a portion of the initial object prediction information to be displayed; receiving adjusted object prediction information pertaining to at least some of the plurality of objects, wherein the adjusted object prediction information is obtained from a user input made via a user interface configured for a user to make annotation adjustments to at least some of the initial object prediction information; and outputting updated object prediction information, wherein the updated object prediction information is based at least in part on the adjusted object prediction information.

Abstract: A technique is disclosed for automating tracking of annotated objects and improves the throughput and efficiency of existing methods while maintaining a degree of accuracy comparable to a human annotator. In particular, the disclosed technique provides an automated annotated object tracking tool that allows machine-learning teams to annotate an object within a frame and have that annotation persist across frames as the annotated object is tracked within a series of frames, still ensuring that every frame is accurately reviewed by a human where high quality annotation is required. This technique incorporates human feedback via a user adjustment that allows the tool to adapt and improve its accuracy in tracking an annotated object across a sequence of frames.

Abstract: Image annotation includes: accessing an image and a plurality of annotation data sets for the image, wherein the plurality of annotation data sets are made by a plurality of contributors, and the image has a plurality of original image channels; aggregating the plurality of annotation data sets to obtain an aggregated annotation data set for the image; and outputting the aggregated annotation data set. Aggregating the plurality of annotation data sets to obtain an aggregated annotation data set for the image includes: generating an additional image channel based at least in part on weight averages of confidence measures of the plurality of contributors; and applying an object detection model to at least a part of the plurality of original image channels and at least a part of the additional image channel to generate the aggregated annotation data set.

Abstract: A technique is disclosed for automating tracking of annotated objects and improves the throughput and efficiency of existing methods while maintaining a degree of accuracy comparable to a human annotator. In particular, the disclosed technique provides an automated annotated object tracking tool that allows machine-learning teams to annotate an object within a frame and have that annotation persist across frames as the annotated object is tracked within a series of frames, still ensuring that every frame is accurately reviewed by a human where high quality annotation is required. This technique incorporates human feedback via a user adjustment that allows the tool to adapt and improve its accuracy in tracking an annotated object across a sequence of frames.

Abstract: Image annotation includes: accessing an image and a plurality of annotation data sets for the image, wherein the plurality of annotation data sets are made by a plurality of contributors, and the image has a plurality of original image channels; aggregating the plurality of annotation data sets to obtain an aggregated annotation data set for the image; and outputting the aggregated annotation data set. Aggregating the plurality of annotation data sets to obtain an aggregated annotation data set for the image includes: generating an additional image channel based at least in part on weight averages of confidence measures of the plurality of contributors; and applying an object detection model to at least a part of the plurality of original image channels and at least a part of the additional image channel to generate the aggregated annotation data set.

Abstract: Image annotation includes: accessing initial object prediction information associated with an image, wherein the initial object prediction information includes a plurality of initial predictions associated with a plurality of objects in the image, including bounding box information associated with the plurality of objects; presenting the image and at least a portion of the initial object prediction information to be displayed; receiving adjusted object prediction information pertaining to at least some of the plurality of objects, wherein the adjusted object prediction information is obtained from a user input made via a user interface configured for a user to make annotation adjustments to at least some of the initial object prediction information; and outputting updated object prediction information, wherein the updated object prediction information is based at least in part on the adjusted object prediction information.

Abstract: A technique is disclosed for automating tracking of annotated objects and improves the throughput and efficiency of existing methods while maintaining a degree of accuracy comparable to a human annotator. In particular, the disclosed technique provides an automated annotated object tracking tool that allows machine-learning teams to annotate an object within a frame and have that annotation persist across frames as the annotated object is tracked within a series of frames, still ensuring that every frame is accurately reviewed by a human where high quality annotation is required. This technique incorporates human feedback via a user adjustment that allows the tool to adapt and improve its accuracy in tracking an annotated object across a sequence of frames.

Abstract: A technique is disclosed for automating tracking of annotated objects and improves the throughput and efficiency of existing methods while maintaining a degree of accuracy comparable to a human annotator. In particular, the disclosed technique provides an automated annotated object tracking tool that allows machine-learning teams to annotate an object within a frame and have that annotation persist across frames as the annotated object is tracked within a series of frames, still ensuring that every frame is accurately reviewed by a human where high quality annotation is required. This technique incorporates human feedback via a user adjustment that allows the tool to adapt and improve its accuracy in tracking an annotated object across a sequence of frames.

Abstract: Systems, methods, and apparatuses are presented for a novel natural language tokenizer and tagger. In some embodiments, a method for tokenizing text for natural language processing comprises: generating from a pool of documents, a set of statistical models comprising one or more entries each indicating a likelihood of appearance of a character/letter sequence in the pool of documents; receiving a set of rules comprising rules that identify character/letter sequences as valid tokens; transforming one or more entries in the statistical models into new rules that are added to the set of rules when the entries indicate a high likelihood; receiving a document to be processed; dividing the document to be processed into tokens based on the set of statistical models and the set of rules, wherein the statistical models are applied where the rules fail to unambiguously tokenize the document; and outputting the divided tokens for natural language processing.

Abstract: An information distribution method includes: gathering all-hazards information into an information exchange from a first information source; gathering all-hazards information into the information exchange from a second information source; distributing the all-hazards information from the information exchange to a first independently-controlled alert network; distributing the all-hazards information from the information exchange to a second independently-controlled alert network.

Abstract: An information distribution method includes: gathering all-hazards information into an information exchange from a first information source; gathering all-hazards information into the information exchange from a second information source; distributing the all-hazards information from the information exchange to a first independently-controlled alert network; distributing the all-hazards information from the information exchange to a second independently-controlled alert network.

Abstract: An information distribution method includes: gathering all-hazards information into an information exchange from a first information source; gathering all-hazards information into the information exchange from a second information source; distributing the all-hazards information from the information exchange to a first independently-controlled alert network; distributing the all-hazards information from the information exchange to a second independently-controlled alert network.

Abstract: An information distribution method includes: gathering all-hazards information into an information exchange from a first information source; gathering all-hazards information into the information exchange from a second information source; distributing the all-hazards information from the information exchange to a first independently-controlled alert network; distributing the all-hazards information from the information exchange to a second independently-controlled alert network.

Abstract: A boat mooring device comprises two PVC pipes pivotally joined at a common midpoint to be selectively positionable from a coextensive linear configuration to an X-shaped configuration. Each pipe has a segment of nylon line extending therethrough and secured therein. The device includes a first pair of line anchoring structures adapted for spaced-apart attachment to a docking structure, and a second pair of line anchoring structures adapted for spaced-apart attachment to a watercraft. In use, the pipes are arranged in an X-shaped configuration and the ends of the nylon lines are attached to the anchoring structures on the dock and watercraft. Each pair of line anchoring structures are spaced apart at a distance approximately equal to half the length of the pipes. In this arrangement, the pipes form an angle of approximately 60° with respect to one another, which provides the optimal configuration for maintaining the moored boat in position.

Abstract: A method and apparatus for reproducing real-world lighting effects in a computer animated object, in which a three dimensional virtual set is created in proportion to the physical set, optical parameters of each physical light source used on the real-world set are input into the model and qualified by effects such as dimming, filtration, diffusion, occlusion and attenuation to produce synthetic lighting of the virtual set. The camera lens transmission characteristics and film speed may also be incorporated into the virtual set parameters.

Abstract: A porous body is formed and joined to the component by sintering a metal powder to the component. The porous body and component are placed into a piston casting die. Molten metal is poured into the die such that the porous body becomes infiltrated with the molten metal and the molten metal is allowed to solidify. The porous body may be formed by sintering loose metal powder to the component, or by forming a separate body by die pressing of metal powder and which body may be subsequently sintered and joined to the component.

Abstract: Pistons and a method for making them are described. The pistons have reinforcing components, such as piston ring groove inserts, incorporated therein. The reinforcements have one or more porous bodies joined thereto, for example, by brazing and which are infiltrated during casting of the piston.