A system generates a plurality of spatial points based on depth measurements of physical objects. The system determines, based on the plurality of spatial points, an occupancy score for each voxel within a plurality of voxels. The system identifies, based on a gaze of the user, a first set of occupied voxels that are in a field of view of the user and a second set of occupied voxels that are outside the field of view of the user. The system updates the occupancy scores of the first set of occupied voxels by temporally decaying one or more of the plurality of spatial points within the first set of occupied voxels. The system maintains the occupancy scores of the second set of occupied voxels. The system detects intrusions in a predefined subspace within a physical space based on the updated occupancy scores of the first set of occupied voxels.

Described herein is a technique for performing operations for a bounding volume hierarchy. The techniques include: for a bounding box with quantized orientation, the bounding box being part of a bounding volume hierarchy, rotating a ray according to the quantized orientation to generate a rotated ray; performing an intersection test against the bounding box with the rotated ray; and according to the results of the intersection test, continuing traversal of the bounding volume hierarchy.

Described herein is a technique for performing ray tracing operations. The technique includes encountering, at a non-leaf node, a pointer to a bottom-level acceleration structure having one or more delta instances; identifying an index associated with the pointer, wherein the index identifies an instance within the bottom-level acceleration structure; and obtaining data for the instance based on the pointer and the index.

A medical system comprises a display system, a user input device, and a control system communicatively coupled to the display system and the user input device. The control system is configured to display image data of an anatomical region via the display system, determine a target location in the anatomical region, and determine an anatomical boundary based on the target location. The anatomical boundary indicates a surface of an anatomical structure in the anatomical region. The control system is further configured to determine a trajectory zone around a path between an exit point and the target location. The control system is further configured to determine a zone boundary based on an intersection of the trajectory zone with the anatomical boundary.

Techniques are disclosed relating to primitive intersection testing for ray tracing in graphics processors. In some embodiments, a graphics processor includes ray intersection circuitry configured to perform an intersection test, which includes to: quantize a first representation of the primitive to generate a reduced-precision interval representation of the primitive, quantize a first representation of the ray to generate a reduced-precision interval representation of the ray, and determine, using interval arithmetic, an initial intersection result based on coordinates of the interval representation of the primitive and coordinates of the interval representation of the ray. The initial intersection result may be a conservative result such that a miss indicated by the initial intersection result is guaranteed not to be a hit for the first representation of the primitive and first representation of the ray. Disclosed techniques may improve performance, reduce power consumption, or both, relative to traditional techniques.

Embodiments of a learning-based excavation planning method are disclosed for excavating rigid objects in clutter, which is challenging due to high variance of geometric and physical properties of objects, and large resistive force during the excavation. A convolutional neural network is utilized to predict a probability of excavation success. Embodiments of a sampling-based optimization method are disclosed for planning high-quality excavation trajectories by leveraging the learned prediction model. To reduce simulation-to-real gap for excavation learning, voxel-based representations of an excavation scene are used. Excavation experiments were performed in both simulation and real world to evaluate the learning-based excavation planners. Experimental results show that embodiments of the disclosed method may plan high-quality excavations for rigid objects in clutter and outperform baseline methods by large margins.

Systems and methods for video presentation and analytics for live sporting events are disclosed. At least two cameras are used for tracking objects during a live sporting event and generate video feeds to a server processor. The server processor is operable to match the video feeds and create a 3D model of the world based on the video feeds from the at least two cameras. 2D graphics are created from different perspectives based on the 3D model. Statistical data and analytical data related to object movement are produced and displayed on the 2D graphics. The present invention also provides a standard file format for object movement in space over a timeline across multiple sports.

Systems and methods for creating a digital model of an indoor space and virtually furnishing the digital model are described. A data collection device photographs or scans an indoor space and produces dimensional data using collected data. The system further includes a pre-processing process, a planes process, a move to origin process, and a segmentation process that analyze and process the dimensional data to create a digital model of the indoor space. The digital model is saved to a database and is viewable on a display and includes a visual scale that corresponds to the spatial dimensions of the indoor space. The system also detects objects present in the indoor space for removal from the digital model. The system also allows creation and placement of visual object representations for display in the digital model.

Techniques for placing and manipulating multiple three-dimensional (3D) models using mobile augmented reality (AR) are described. One technique includes receiving a first request to initialize an AR simulation of a first product for sale within a physical environment. In response to the first request, a first 3D model of the first product for sale is rendered onto the screen. After rendering the first 3D model, a second request to visualize a second product for sale within the physical environment is received during the AR simulation of the first product for sale. In response to the second request, a second 3D model of the second product for sale is rendered onto the screen with the first 3D model.

A physics engine executed on a processor to simulate rigid body dynamics of a simulated physical system using an inertia scaling function is provided. The physics engine may be configured to iteratively loop through a collision detection phase, an iterative solving phase, updating phase, and display phase. The physics engine may further be configured to determine a neighboring body weighting value for one or more of the plurality of bodies, and determine an inertia scaling value for the one or more of the plurality of bodies based on the neighboring body weighting value for that body. The physics engine may further be configured to scale an inertia value for a body of that colliding pair of bodies based on the inertia scaling value for the iterative solving phase.