Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating output sequences using auto-regressive decoder neural networks. In particular, during generation, adaptive early exiting is used to reduce the time required to generate the output sequence.

Abstract: A method, computer system, and a computer program product for projecting a 3D model defined by x, y, z coordinates onto the surface of a 2D image defined by u, v coordinates is provided. The present invention may include receiving a 3D model having a plurality of polygons, wherein certain edges are marked as seams. The present invention may include receiving input from a user, wherein the input comprises painting one or more parts of the 3D model in different colors, wherein the colors correspond with a weight of the area painted. The present invention may include unwrapping, by a processor, a 2D texture from the 3D model using a projection algorithm. The present invention may include generating a rectangular boundary around each island. The present invention may include scaling each island according to a gradient score.

Abstract: Disclosed herein are image-based object recognition method and system by and in which a learning server performs image-based object recognition based on the learning of environment variable data. The image-based object recognition method includes: receiving an image acquired through at least one camera, and segmenting the image on a per-frame basis; primarily learning labeling results for one or more objects in the image segmented on a per-frame basis; performing primary reasoning by performing object detection in the image through a model obtained as a result of the primary learning; performing data labeling based on the results of the primary reasoning, and performing secondary learning with weights allocated to respective boxes obtained by the primary reasoning and estimated as object regions; and finally detecting one or more objects in the image through a model generated as a result of the secondary learning.

Abstract: A computer-implemented method according to one embodiment includes performing object detection on a frame to identify a set of objects within the frame, and determining a location of each object of the set of objects within the frame. The method further includes constructing a code tree structure for the frame. The code tree structure defines the determined locations. A rule defining modifications to make to at least some of the objects is received and the code tree structure is crawled. The rule is applied to the objects during the crawling. The method further includes layering the modifications defined in the rule on the set of objects within the frame. A computer program product according to another embodiment includes a computer readable storage medium having program instructions embodied therewith. The program instructions are readable and/or executable by a computer to cause the computer to perform the foregoing method.

Abstract: Several heart wall remodeling devices and methods are disclosed. One heart wall remodeling device includes a trimming tool for trimming a line within a patient. The trimming tool includes an actuator for manipulation by a user, a moveable inner shaft coupled to the actuator, and an outer sleeve coupled to the inner shaft. The outer shaft includes a backstop disposed along an inner wall of the outer sleeve and a channel disposed between a first end and a second end of the outer sleeve. The inner shaft includes an inner shaft, a compressible member, and a cutter.

Abstract: Disclosed herein are image-based object recognition method and system by and in which a learning server performs image-based object recognition based on the learning of environment variable data. The image-based object recognition method includes: receiving an image acquired through at least one camera, and segmenting the image on a per-frame basis; primarily learning labeling results for one or more objects in the image segmented on a per-frame basis; performing primary reasoning by performing object detection in the image through a model obtained as a result of the primary learning; performing data labeling based on the results of the primary reasoning, and performing secondary learning with weights allocated to respective boxes obtained by the primary reasoning and estimated as object regions; and finally detecting one or more objects in the image through a model generated as a result of the secondary learning.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating output sequences using auto-regressive decoder neural networks. In particular, during generation, adaptive early exiting is used to reduce the time required to generate the output sequence.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating output sequences using auto-regressive decoder neural networks. In particular, during generation, adaptive early exiting is used to reduce the time required to generate the output sequence.

Abstract: A dynamic asset spawning engine is provided in a data processing system for dynamically spawning assets in a video game based on one or more news feeds. The dynamic asset spawning engine semantically tags a library of video game assets with game tag descriptors. The dynamic asset spawning engine parses and processes the one or more news feeds to identify supplemental tag descriptors and spawning parameters for one or more video game assets in the library of video game assets. The dynamic spawning engine assigns the identified supplemental tag descriptors and spawning parameters to the one or more video game assets. The video game spawns the one or more game assets in the video game in accordance with the supplemental tag descriptors and spawning parameters.

Abstract: Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: establishing a relationship graph for a certain user, the relationship graph having a primary node that specifies the certain user and a plurality of lower order nodes, wherein the primary node and at least one lower order node of the plurality of lower order nodes is connected to at least one other node by an edge so that there are defined a plurality of edges; generating a prioritized list of individuals to provide assistance to the certain user in respect to the certain task based on populated edge weight scores for edges of the relationship graph; and providing one or more output based on the prioritized list.

Abstract: A vascular occluding device for modifying blood flow in a vessel, while maintaining blood flow to the surrounding tissue. The occluding device includes a flexible, easily compressible and bendable occluding device that is particularly suited for treating aneurysms in the brain. The neurovascular occluding device can be deployed using a micro-catheter. The occluding device can be formed by braiding wires in a helical fashion and can have varying lattice densities along the length of the occluding device. The occluding device could also have different lattice densities for surfaces on the same radial plane.

Abstract: The system includes an optical resonator, a mount, and a fastener. The optical resonator is comprised of a material with a horizontal plane symmetry. The optical resonator includes a horizontal plane protrusion for mounting. The horizontal plane protrusion includes discrete resonator rotational orientation positions. The mount comprises mounting legs compatible with the horizontal plane symmetry. The mount includes discrete mount rotational orientation positions that correspond to the discrete resonator rotation orientation positions at a plurality of rotational angles. The fastener secures the horizontal plane protrusion of the optical resonator to the mount.

Abstract: The invention relates to a compacted and densified metal material having one or more phases formed of an agglomerate of grains, the cohesion of the material being provided by bridges formed between grains, said material having a relative density higher than or equal to 95% and preferably higher than or equal to 98%.

Abstract: A system and method for identifying a subject using imaging are provided. In some aspects, the method includes receiving an image depicting a subject to be identified, and applying a trained Disentangled Representation learning-Generative Adversarial Network (DR-GAN) to the image to generate an identity representation of the subject, wherein the DR-GAN comprises a discriminator and a generator having at least one of an encoder and a decoder. The method also includes identifying the subject using the identity representation, and generating a report indicative of the subject identified.

Abstract: Aspects of the present disclosure relate to a method of processing an input image of a culture device for a condition. The method can include receiving the input image and classifying the input image with a trained machine learning model that is configured to be trained on a training set of images having the condition. The method can include determining that the condition exists in the input image based on the classification and performing at least one action in response to the determination that the condition exists.

Abstract: Aspects of the present disclosure relate to a method of colony enumeration. The method includes identifying colony forming units of microorganisms in a combined image using a pretrained deep learning model on a colony enumeration device. The method can include providing a plurality of identification characteristics of the colony forming units to an interaction component such that the interaction component can project at least some of the plurality of identification characteristics onto the combined image.

Abstract: An elbow joint connector includes an elbow portion and a locking mechanism detachably coupled to a distal end of the elbow portion. The locking mechanism including a pair of locking members translatable relative to one another along an interface transverse to a longitudinal axis of the elbow joint connector. Divergent translation of the pair of locking members relative to one another along the interface increases a portion of an outer periphery of the locking mechanism to lock the elbow portion to an adjoining member. The elbow joint connector is configured for detachable coupling to an adjoining elbow joint connector and pivotally rotatable relative thereto from a first configuration in which respective center axes of the connectors form an acute angle to a second configuration in which the respective center axes are collinear.

Abstract: A method, computer system, and a computer program product for determining locations for seams on a 3D model of an object is provided. The present invention may include training an artificial intelligence model using a set of training data. The present invention may include generating a first model for the object using a shrink wrap method. The present invention may include generating a second model for the object using a decimation method. The present invention may include comparing the object to objects in the set of training data to identify an object in the training data having a similar shape. The present invention may include identifying the object by determining if the object fits in between the first and second models. The present invention may lastly include projecting seams onto a model of the object using the trained artificial intelligence model.

Abstract: A method, computer system, and a computer program product for projecting a 3D model defined by x, y, z coordinates onto the surface of a 2D image defined by u, v coordinates is provided. The present invention may include receiving a 3D model having a plurality of polygons, wherein certain edges are marked as seams. The present invention may include receiving input from a user, wherein the input comprises painting one or more parts of the 3D model in different colors, wherein the colors correspond with a weight of the area painted. The present invention may include unwrapping, by a processor, a 2D texture from the 3D model using a projection algorithm. The present invention may include generating a rectangular boundary around each island. The present invention may include scaling each island according to a gradient score.

Abstract: A method is provided. The method may include, in response to electronically receiving on a first computing device a navigation file from a secondary computing device, generating a graphical navigation guide for a user interface (UI) based on the navigation file, wherein generating the graphical navigation guide comprises generating computer operations for the first computing device corresponding to the sequence of computer operations from the navigation file. The method may further include, based on the generated computer operations, executing the graphical navigation guide on the UI associated with the first computing device, wherein executing the graphical navigation guide comprises displaying a screen and a UI element corresponding to the sequence of computer operations, and wherein displaying the UI element comprises rendering an overlay on the UI element that highlights the UI element on the displayed screen and instructs a user to perform an input action on the UI element.