Microsoft Technology Licensing, LLC

Cccc Highway Consultants Co. Ltd.

China Communications Construction Company Limited

Ke Deng

Baining Guo

Lifeng Wang

Zhuowen Tu

Eric David Heutchy

Joshua William Buckman

Qiang Wang

Jinyu Li

Xiaodong Wang

Wei Lin

Tommer Leyvand

Szymon Piotr Stachniak

Xiaodong Liu

Zhigang Zhang

Hai Ji

Baoyuan Wang

Warren Lee Burch

Matthew Gordon

Ming Lin

Wei Shen

Chen Li

Qian Cheng

Roger John Perkins

Haiqing Yin

Jibing Gao

Aaron Blake Stover

Yonggang Lv

Lingfeng Liu

Huaiping Su

A computing system configured to execute instructions for a first graphical processing unit (GPU) on a second GPU is provided. The computing system may include the second GPU and a processor. The processor may be configured to receive second GPU state data that indicates one or more global properties of the second GPU. The processor may be further configured to receive one or more binary instructions for texture operation configured to be executed on the first GPU. Based on the second GPU state data, the processor may be further configured to apply a texture value patch to the one or more binary instructions. Applying the texture value patch may translate the one or more binary instructions into one or more translated binary instructions configured to be executed on the second GPU.

Texture value patch used in GPU-executed program sequence cross-compilation

The present application discloses a final joint of an immersed tunnel, a prefabrication method and an installation method, wherein the final joint includes two end surfaces connected with installed adjacent tube sections; the two end surfaces are both tilted surfaces, so that the longitudinal profile of the final joint along an installation direction is of an inverted trapezoid structure; and the final joint further may be of a structure with a tube section I and a tube section II which are connected with each other. The final joint of the immersed tunnel is simple in structure, convenient to control and relatively high in precision, thereby reducing lots of open sea diving work and lowering a risk of installation quality defects; as prefabrication procedures are simple, the final joint may be prefabricated in a land factory and then transported to the site, thereby reducing influence of weather conditions on construction; a body structure of the final joint is prefabricated in the factory, and then the overall final joint is transported to the site for installation; water stop systems realize quick water stop, thus forming a dry construction environment; and therefore, the influence of weather and tidal current conditions on a project may be reduced, and a quality risk may be lowered.

Final joint of immersed tunnel as well as prefabrication method and installation method

Low-latency gesture detection is described, for example, to compute a gesture class from a live stream of image frames of a user making a gesture, for example, as part of a natural user interface controlling a game system or other system. In examples, machine learning components are trained to learn gesture primitives and at test time, are able to detect gestures using the learned primitives, in a fast, accurate manner. For example, a gesture primitive is a latent (unobserved) variable features of a subset of frames from a sequence of frames depicting a gesture. For example, the subset of frames has many fewer frames than a sequence of frames depicting a complete gesture. In various examples gesture primitives are learnt from instance level features computed by aggregating frame level features to capture temporal structure. In examples frame level features comprise body position and body part articulation state features.

Low-latency gesture detection

Resource processing during run time can be performed asynchronously from emulation of an application by a central processing unit. For example, an emulator can include a main processing thread that performs emulation processes. In response to encountering a shader, or other resource, to be processed, the emulator can invoke a separate asynchronous thread to perform such processing. Processed resources, such as translated shaders and generated textures, can be stored in a cache. In response to a command that uses a resource, such as a draw command that invokes a shader or other resource, the emulator can use the processed resource in the cache. If the processed resource is not in the cache, the emulator can skip processing the command that uses the resource. If processed resources can be obtained from other sources and loaded in the cache, processing of resources by the emulator can be eliminated.

Asynchronous translation of computer program resources in graphics processing unit emulation

Embodiments are disclosed that relate to the correction of an estimated pose determined from depth image data. One disclosed embodiment provides, on a computing system, a method of obtaining a representation of a pose of articulated object from image data capturing the articulated object. The method comprises receiving the depth image data, obtaining an initial estimated skeleton of the articulated object from the depth image data, applying a random forest subspace regression function to the initial estimated skeleton, and determining the representation of the pose based upon a result of applying the random forest subspace regression to the initial estimated skeleton.

Estimated pose correction

Emulating legacy hardware using IEEE 754 compliant hardware is disclosed herein. In some aspects, the emulation includes locating an instruction that includes NaN (not a number) as at least one of an operand or a resultant. The emulation adjusts the resultant of the instruction, via additional code, to produce a final resultant of non-compliant (legacy) hardware. Legacy software, which was written in anticipation of processing by legacy hardware, may then be processed using compliant hardware.

Emulating legacy hardware using IEEE 754 compliant hardware

A software-implemented transform and lighting module and pipeline designed and optimized for embedded platforms (such as mobile computing devices). The transform and lighting module and pipeline includes a number of features that make it well-suited for use on embedded devices. These features include a single streamline branched architecture that allows efficient processing on a CPU of an embedded device and saves computational time. This architecture is facilitated by use of a vertex cache that stores vertices as needed to avoid duplication in processing of the vertices. A culling feature culls vertices before lighting instead of lighting all vertices. A back face culling technique examines each of the vertices to determines whether a back face of a triangle is formed. If so, then the vertex is culled. A second technique involved determining whether a vertex is outside of one view frustum clip plane. If so, then the vertex is culled.

Software-implemented transform and lighting module and pipeline for graphics rendering on embedded platforms using a fixed-point normalized homogenous coordinate system

A software-implemented graphics rendering system and method designed and optimized for embedded devices (such as mobile computing devices) using fixed-point operations including a variable-length fixed point representation for numbers and a normalized homogenous coordinates system for vector operations. The graphics rendering system and method includes a fixed-point mathematics library and graphics functions that includes optimized basic functions such as addition, subtraction, multiplication, division, all vertex operations, matrix operations, transform functions and lighting functions, and graphics functions. The mathematical library and graphics functions are modified and optimized by using a variable-length fixed-point representation and a normalized homogenous coordinate system (NHCS) for vector operations.

Optimized fixed-point mathematical library and graphics functions for a software-implemented graphics rendering system and method using a normalized homogenous coordinate system

Growing community of inventors

Sammamish, WA, United States of America

Ke Deng