Abstract: Various embodiments set forth systems and techniques for generating seams for a 3D model. The techniques include generating, based on the 3D model, one or more inputs for one or more trained machine learning models; providing the one or more inputs to the one or more trained machine learning models; receiving, from the one or more trained machine learning models, seam prediction data generated based on the one or more inputs; and placing one or more predicted seams on the 3D model based on the seam prediction data.

Abstract: Techniques and systems for computer aided design of physical structures using an object splitting design process that optimize manufacturing efficiency are described. A described technique includes splitting, by a computer program, a three dimensional model of an initial object into an up skin region and a down skin region based on an analysis of draft angles along at least one surface of the three dimensional model of the initial object. The technique further includes producing, by the computer program, first and second three dimensional models of respective first and second objects in accordance with the up skin region and the down skin region to create separate first and second physical structures using one or more computer-controlled manufacturing systems such that the first and second physical structures are combinable to form a combined physical structure.

Abstract: A method and system provide the ability to design a terrain surface. A triangular surface mesh in three-dimensions representative of an existing surface is obtained. One or more constraints to control the triangular surface mesh are specified. The specifying includes defining multiple basic grading element constraints that are constraints on surface points of the existing surface, and combining such constraints into a complex grading element constraint that matches a real world grading behavior. Drainage for the triangular surface mesh is automatically determined based on the complex grading element constraint.

Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for providing editable keyframe-based animation data for applying to a character to animate motion of the character in three-dimensional space. Three-dimensional motion data is constructed from two-dimensional videos. The three-dimensional motion data represents movement of people in the two-dimensional videos and includes, for each person, a root of a three-dimensional skeleton of the person. The three-dimensional skeleton comprises multiple three-dimensional poses of the person during at least a portion of frames of a video from the two-dimensional videos. The three-dimensional motion data is converted into editable keyframe-based animation data in three-dimensional space and provided to animate motion.

Abstract: In various embodiments, a gradient modeling application automatically generates designs of three-dimensional (3D) objects. The gradient modeling application generates a set of points based on a resolution and a perimeter. The gradient modeling application computes a set of displacement values based on the set of points, a first two-dimensional (2D) border, and a first displacement parameter that is associated with the first 2D border. Based on the set of displacement values, the gradient modeling application generates a 3D object design.

Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for designing three dimensional lattice structures include, in one aspect, a method including: obtaining a mechanical problem definition including a 3D model of an object; generating a numerical simulation model for the 3D model of the object using one or more loading cases and one or more isotropic solid materials identified as a baseline material model for a design space; predicting performance of different lattice settings in different orientations in the design space using a lattice structural behavior model in place of the baseline material model in the numerical simulation model; and presenting a set of lattice proposals for the design space based on the predicted performance of the different lattice settings in the different orientations; wherein the lattice structural behavior model has been precomputed for the different lattice settings, which are generable by the 3D modeling program.

Abstract: A hybrid workstation enables a virtual reality (VR) interface, a traditional (TD) interface, and transitions between the interfaces. The VR interface comprises three-dimensional (3D)-based software and hardware components. The TD interface comprises two-dimensional (2D)-based software and hardware components. The state of the hybrid workstation is defined by three parameters comprising interface (VR interface or TD interface), position (seated or standing), and movement (stationary or room-scale). The hybrid workstation detects a transition from a current state to a next state upon determining that any of the three parameters have changed. The hybrid workstation then determines a transition response based on the particular transition that is detected. The transition response comprises a set of operations that are performed on the VR interface and/or the TD interface that mitigate the disruption and inefficiency caused when the particular transition occurs.

Abstract: A finger device initiates actions on a computer system when placed in contact with a surface. The finger device includes instrumentation that captures images and gestures. When in contact with a surface, the finger device captures images of the surface and gestures made on the surface. The finger device also transmits the images and gesture data to the computer system. An application on the computer system matches the images received from the finger device to a representation of the surface, identifies an action associated with the surface representation and gesture, and executes the action. Instrumenting the finger instead of the surface, allows a user to configure virtually any surface to accept touch input.

Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for computer aided design of physical structures include, in at least one aspect, a fully automatic method of converting a generative design into an editable, watertight B-Rep by leveraging the generative solver input and representation to: (1) embed the exact input solid boundary surfaces where the design coincides with the input, (2) approximate everywhere else the design boundary with globally smooth, editable “organic” surfaces, and (3) join all surfaces to form a generative design output B-Rep.

Abstract: A design analysis engine analyzes a CAD assembly to generate a topological model. The topological model is a graph of nodes coupled together via a set of edges. Each node represents a different CAD model in the CAD assembly and each edge represents a different connection between two such CAD models. The design analysis engine also analyzes the CAD assembly to generate a data model corresponding to the topological model. The data model is a structured dataset that includes component entries and connection entries. A component entry includes design data associated with a CAD model in the CAD assembly and a connection entry includes design data associated with a physical or logical connection between two or more such CAD models. A user interacts with the topological model to navigate the CAD assembly, obtain CAD model data, and initiate automatically-performed design tasks.

Abstract: An urban design pipeline automatically generates design options for an urban design project. The urban design pipeline includes a geometry engine and an evaluation engine. The geometry engine analyzes design criteria, design objectives, and design heuristics associated with the urban design project and then generates numerous candidate designs. The design criteria specify a property boundary associated with a region of land to be developed. The design objectives indicate a specific type of topology that is derived from an existing urban layout. The design heuristics include different sets of construction rules for generating designs with specific types of topologies. The geometry engine generates candidate designs that conform to the property boundary and have topological characteristics in common with the existing urban layout.

Abstract: A method, apparatus, system, and computer program product provide the ability to dynamically generate a digital building information model. Design data for various designs is received. The design data for each design is encoded into a graph. A knowledge base is maintained and defines a model of design intent while processing and storing the graph. First user input of a goal or constraint is received. The knowledge base generates solutions based on the input. Second user input based on the solutions is received and used to iteratively train the knowledge base. The solutions are then output.

Abstract: Various embodiments set forth systems and techniques for assessing sustainability of a building design. The techniques include receiving a building design for a budding and sustainability certification data describing a plurality of sustainability certification systems; selecting one or more particular sustainability certification systems from the plurality of sustainability certification systems; computing a respective sustainability certification value for each sustainability certification system of the one or more particular sustainability certification systems; and performing one or more building design operations based on sustainability certification system values for the one or more particular sustainability certification systems.

Abstract: In various embodiments, a stylization application generates designs that reflect stylistic preferences. In operation, the stylization application computes characterization information based on a first design and a trained machine-learning model that maps one or more designs to characterization information associated with one or more styles. The stylization application then computes a style score based on the characterization information and a target style that is included in the one or more styles. Subsequently, the stylization application generates a second design based on the style score, where the second design is more representative of the target style than the first design. Advantageously, because the stylization application can substantially increase the number of designs that can be generated based on the target style in a given amount of time, relative to more manual prior art techniques, the overall quality of the design ultimately selected for production can be improved.

Abstract: Techniques disclosed herein combine computer vision with eye tracking by identifying, via computer vision, safety hazards in a video captured by a camera mounted on a pair of safety glasses, and generating an alert if a user wearing the safety glasses has not (recently) noticed the hazard. Whether the user has noticed the hazard is determined based on eye tracking information extracted from a video captured by another camera that is mounted on the safety glasses and points toward the user. As a result, safety hazards may be automatically detected and reported to the user. In addition, only those hazards that have not been (recently) noticed by the user cause an alert to be generated, so the user is not distracted with unnecessary notifications.

Abstract: A method and system provide a workflow for extracting building information model (BIM) elements for a floor plan drawing. A design drawing area of an image of the floor plan drawing is determined. Elements are extracted from the design drawing area. A synthetic floor plan design drawing dataset is obtained with known synthetic symbol labels and known synthetic symbol locations. Based on the extracted elements and the synthetic floor plan design drawing dataset, a symbol represented by the extracted elements is detected. Based on the symbol, a building information model (BIM) element is fetched and placed in the floor plan drawing.

Abstract: One embodiment of the invention disclosed herein provides techniques for voxelizing a mesh representation associated with a three-dimensional model to generate a volumetric model. A model filling engine associated with a voxelization system identifies a first voxel included in a voxel grid array that intersects with the mesh representation. The model filling engine selects a second voxel at an exterior boundary of the voxel grid array that is not identified as a boundary voxel. The model filling engine marks the second voxel as an exterior voxel. The model filling engine marks all unmarked voxels that are adjacent to the second voxel as exterior voxels. The model filling engine marks all remaining voxels as interior voxels. A model finishing engine associated with the voxelization system generates a volumetric model based at least in part on the first voxel.

Abstract: A design application is configured to generate a latent space representation of a fleet of pre-existing vehicles. The design application encodes vehicle designs associated with the fleet of pre-existing vehicles into the latent space representation to generate a first latent space location. The first latent space location represents the characteristic style associated with the fleet of pre-existing vehicles. The design application encodes a sample design provided by a user into the latent space representation to produce a second latent space location. The design application then determines a distance between the first latent space location and the second latent space location. Based on the distance, the design application generates a style metric that indicates the aesthetic similarity between the sample design and the vehicle designs associated with the fleet of pre-existing vehicles. The design application can also generate new vehicle designs based on the latent space representation and the sample design.

Abstract: One embodiment of the present application sets forth a computer-implemented method for remotely monitoring performance of a task in real time, comprising receiving, in real time from a first student computing device, first student data that includes a first set of actions performed in a first student application instance executing on the first student device, comparing a first student action included in the first set of actions to a first target action, in response to comparing the first student action to the first target action, identifying a first issue indicator associated with the first student action, determining a first remedy associated with at least one of the first target action and the first issue indicator, and performing a first remediation action associated with the first remedy.

Abstract: In various embodiments, a pattern-based recommendation subsystem automatically recommends workflows for software-based tasks. In operation, the pattern-based recommendation subsystem computes an expected distribution of frequencies across command patterns based on different distributions of frequencies across the command patterns. The expected distribution of frequencies is associated with a target user, and each different distribution of frequencies is associated with a different user. The pattern-based recommendation subsystem then applies a set of commands associated with the target user to a trained machine-learning model to determine a target distribution of weights applied to a set of tasks. Subsequently, the pattern-based recommendation subsystem determines a training item based on the expected distribution of frequencies and the target distribution of weights. The pattern-based recommendation subsystem generates a recommendation that specifies the training item.