DOCSLIB.ORG

Khronos Openvx Webinar Sep2017

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Khronos Openvx Webinar Sep2017 The OpenVX Standard for Portable, Efficient Vision Processing Frank Brill Radha Giduthuri Tom e r Schwartz Cadence AMD Intel September 2017 © Copyright Khronos Group 2017 - Page 1 Khronos Connects Software to Silicon Industry Consortium creating OPEN STANDARD APIs for hardware acceleration Any company is welcome – one company one vote ROYALTY-FREE specifications Conformance Tests and Adopters State-of-the art IP framework protects Software Programs for specification integrity members AND the standards and cross-vendor portability Low-level silicon APIs needed on International, non-profit organization every platform: graphics, parallel Membership and Adopters fees cover compute, vision, neural networks, Silicon operating and engineering expenses augmented and virtual reality… Strong industry momentum 100s of man years invested by industry experts Well over a BILLION people use Khronos APIs Every Day… © Copyright Khronos Group 2017 - Page 2 Khronos Open Standards 3D for the Web - VR/AR and games in-browser - Efficiently delivering runtime 3D assets Real-time 2D/3D - Virtual and Augmented Reality displays - Cross-platform gaming and UI - CAD and Product Design Vision and Neural Networks - Tracking and odometry - Scene analysis/understanding - Neural Network inferencing Parallel Computation - Machine Learning acceleration - Embedded vision processing - High Performance Computing (HPC) © Copyright Khronos Group 2017 - Page 3 OpenVX Vision Engines Wide range of vision hardware architectures Middleware OpenVX provides a high-level Graph-based abstraction -> Applications Enables Graph-level optimizations! Can be implemented on almost any hardware or processor! -> Portable, Efficient Vision Processing! Software GPU Portability DSP Hardware X100 Dedicated Hardware Vision DSPs X10 Vision GPU Node Compute Vision Vision Node Node Multi-core CNN CPU Nodes Power Efficiency X1 Computation Flexibility Vision Processing Graph © Copyright Khronos Group 2017 - Page 4 Neural Net Workflow with Khronos Standards Neural Network Training Third Vision/AI Party Applications Frameworks Tools Datasets Trained Vision and Neural Network Network Inferencing Runtime Network Model Structure Desktop and Cloud Hardware Embedded/Mobile Vision/Inferencing Embedded/Mobile Vision/Inferencing Embedded/MobileHardware Vision/Inferencing Embedded/Mobile/HardwareDesktop/Cloud Hardware OpenVX is an open, royalty-free standard for cross platform acceleration Vision/Inferencing Hardware of computer vision and neural network applications. GPU DSP CPU NNEF is a Cross-vendor Neural Net file format Custom FPGA Encapsulates network formal semantics, structure, data formats, commonly-used operations (such as convolution, pooling, normalization, etc.) © Copyright Khronos Group 2017 - Page 5 Problem: Neural Network Fragmentation Neural Network Inferencing Fragmentation toll on Applications Inference Engine 1 Hardware 1 Vision/AI Hardware 2 Application Inference Engine 2 Hardware 3 Every Application Needs know Inference Engine 3 about Every Accelerator API Neural Network Training and Inferencing Fragmentation NN Authoring Framework 1 Inference Engine 1 NN Authoring Framework 2 Inference Engine 2 NN Authoring Framework 3 Inference Engine 3 Every Tool Needs an Exporter to Every Accelerator © Copyright Khronos Group 2017 - Page 6 Solving Neural Network Fragmentation Neural Network Inferencing Fragmentation toll on Applications Inference Engine 1 Hardware 1 Vision/AI Hardware 2 Application Inference Engine 2 Inference Engine 3 Hardware 3 Neural Network Training and Inferencing Fragmentation NN Authoring Framework 1 Inference Engine 1 NN Authoring Framework 2 Inference Engine 2 NN Authoring Framework 3 Inference Engine 3 Optimization and processing tools © Copyright Khronos Group 2017 - Page 7 OpenVX in a nut-shell • OpenVX contains - a set of memory objects that abstract the physical memory - a library of predefined and customizable vision and neural network functions Image processing and image filtering Feature detectors Neural networks Control flow - a graph-based execution model to combine function enabling both task and data-independent execution • OpenVX is defined as a C API - object-oriented design - synchronous and asynchronous execution model - extend functionality using enum and callback functions © Copyright Khronos Group 2017 - Page 8 OpenVX - Graph-Level Abstraction • OpenVX developers express a graph of image operations (‘Nodes’) - Using a C API • Nodes can be executed on any hardware or processor coded in any language - Implementers can optimize under the high-level graph abstraction • Graphs are the key to run-time power and performance optimizations - E.g. Node fusion, tiled graph processing for cache efficiency etc. Pyr Camera OpenVX Nodes t Rendering Input Output RGB YUV Gray Array of Frame Color Frame Channel Frame Image Optical Keypoints Harris Conversion Extract Pyramid Flow Track Array of Features Ftr OpenVX Graph t-1 Feature Extraction Example Graph © Copyright Khronos Group 2017 - Page 9 OpenVX Efficiency through Graphs.. Graph Memory Kernel Data Scheduling Management Fusion Tiling Split the graph Execute a sub- Reuse execution across Replace a sub- graph at tile pre-allocated the whole graph with a granularity memory for system: CPU / single faster node instead of image multiple GPU / dedicated granularity intermediate data HW Faster execution Less allocation overhead, Better memory Better use of or lower power more memory for locality, less kernel data cache and consumption other applications launch overhead local memory © Copyright Khronos Group 2017 - Page 10 Simple Edge Detector in OpenVX vx_graph g = vxCreateGraph(); vx_image input = vxCreateImage(1920, 1080); Declare Input and Output Images vx_image output = vxCreateImage(1920, 1080); vx_image horiz = vxCreateVirtualImage(g); Declare Intermediate Images vx_image vert = vxCreateVirtualImage(g); vx_image mag = vxCreateVirtualImage(g); vxSobel3x3Node(g, input, horiz, vert); Construct the Graph topology vxMagnitudeNode(g, horiz, vert, mag); vxThresholdNode(g, mag, THRESH, output); Compile the Graph Execute the Graph status = vxVerifyGraph(g); h status = vxProcessGraph(g); i S M m T o v © Copyright Khronos Group 2017 - Page 11 OpenVX Evolution Conformant Conformant Implementations Implementations New Functionality Conditional node execution Feature detection Classification operators Expanded imaging operations New Functionality Extensions Under Discussion Neural Network Acceleration New Functionality NNEF Import Expanded Nodes Functionality Graph Save and Restore 16-bit image operation Enhanced Graph Framework Programmable user AMD OpenVX Tools kernels with accelerator offload - Open source, highly optimized Safety Critical for x86 CPU and OpenCL for GPU OpenVX 1.1 SC for - “Graph Optimizer” looks at Streaming/pipelining safety-certifiable systems entire processing pipeline and removes, replaces, merges functions to improve performance OpenVX and bandwidth Roadmap - Scripting for rapid prototyping, OpenVX 1.2 without re-compiling, at production performance levels Spec released May 2017 http://gpuopen.com/compute-product/amd-openvx/ OpenVX 1.0 OpenVX 1.1 Spec released October 2014 Spec released May 2016 © Copyright Khronos Group 2017 - Page 12 OpenVX SC - Safety Critical Vision Processing • OpenVX 1.1 - based on OpenVX 1.1 main specification - Enhanced determinism - Specification identifies and numbers requirements • MISRA C clean per KlocWorks v10 • Divides functionality into “development” and “deployment” feature sets - Adds requirement to support import/export extension OpenVX SC Verify Binary Import OpenVX SC Development Feature format Deployment Feature Set Set (Create Graph) Export (Execute Graph) Entire graph creation API Implementation No graph creation API dependent format © Copyright Khronos Group 2017 - Page 13 Benefits of Import-Export • Minimize overheads on the target system - Faster setup-time - Simpler OpenVX runtime - Simpler certification (ISO 26262, etc..) • Enable more graph optimizations when offline - No timing/resource constraints - Example: generate & compile specialized code for graphs © Copyright Khronos Group 2017 - Page 14 How OpenVX Compares to Alternatives Open standard API designed to be Community-driven, Open standard API designed to be Governance implemented and shipped by IHVs open source library implemented and shipped by IHVs Programming Graph defined with C API and then Immediate runtime function calls – Explicit kernels are compiled and Model compiled for run-time execution reading to and from memory executed via run-time API Built-in Vision Small but growing Vast. None. User programs their own or Functionality set of popular functions Mainly on PC/CPU call vision library over OpenCL Target Any combination of processors or Any heterogeneous combination of Mainly PCs and GPUs Hardware non-programmable hardware IEEE FP-capable processors Optimization Pre-declared graph enables Each function reads/writes memory. Any execution topology can be Opportunities significant optimizations Power performance inefficient explicitly programmed Implementations must pass Extensive Test Suite but Implementations must pass Conformance conformance to use trademark no formal Adopters program conformance to use trademark All core functions must be available Available functions vary depending on All core functions must be available Consistency in conformant implementations implementation / platform in all conformant implementations © Copyright Khronos Group 2017 - Page 15 Layered Vision/
Load more

Recommended publications
  • Introduction to the Vulkan Computer Graphics API
    1 Introduction to the Vulkan Computer Graphics API Mike Bailey mjb – July 24, 2020 2 Computer Graphics Introduction to the Vulkan Computer Graphics API Mike Bailey [email protected] SIGGRAPH 2020 Abridged Version This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License http://cs.oregonstate.edu/~mjb/vulkan ABRIDGED.pptx mjb – July 24, 2020 3 Course Goals • Give a sense of how Vulkan is different from OpenGL • Show how to do basic drawing in Vulkan • Leave you with working, documented, understandable sample code http://cs.oregonstate.edu/~mjb/vulkan mjb – July 24, 2020 4 Mike Bailey • Professor of Computer Science, Oregon State University • Has been in computer graphics for over 30 years • Has had over 8,000 students in his university classes • [email protected] Welcome! I’m happy to be here. I hope you are too ! http://cs.oregonstate.edu/~mjb/vulkan mjb – July 24, 2020 5 Sections 13.Swap Chain 1. Introduction 14.Push Constants 2. Sample Code 15.Physical Devices 3. Drawing 16.Logical Devices 4. Shaders and SPIR-V 17.Dynamic State Variables 5. Data Buffers 18.Getting Information Back 6. GLFW 19.Compute Shaders 7. GLM 20.Specialization Constants 8. Instancing 21.Synchronization 9. Graphics Pipeline Data Structure 22.Pipeline Barriers 10.Descriptor Sets 23.Multisampling 11.Textures 24.Multipass 12.Queues and Command Buffers 25.Ray Tracing Section titles that have been greyed-out have not been included in the ABRIDGED noteset, i.e., the one that has been made to fit in SIGGRAPH’s reduced time slot.
    [Show full text]
  • Standards for Vision Processing and Neural Networks
    Standards for Vision Processing and Neural Networks Radhakrishna Giduthuri, AMD [email protected] © Copyright Khronos Group 2017 - Page 1 Agenda • Why we need a standard? • Khronos NNEF • Khronos OpenVX dog Network Architecture Pre-trained Network Model (weights, …) © Copyright Khronos Group 2017 - Page 2 Neural Network End-to-End Workflow Neural Network Third Vision/AI Party Applications Training Frameworks Tools Datasets Trained Vision and Neural Network Network Inferencing Runtime Network Model Architecture Desktop and Cloud Hardware Embedded/Mobile Embedded/MobileEmbedded/Mobile Embedded/Mobile/Desktop/CloudVision/InferencingVision/Inferencing Hardware Hardware cuDNN MIOpen MKL-DNN Vision/InferencingVision/Inferencing Hardware Hardware GPU DSP CPU Custom FPGA © Copyright Khronos Group 2017 - Page 3 Problem: Neural Network Fragmentation Neural Network Training and Inferencing Fragmentation NN Authoring Framework 1 Inference Engine 1 NN Authoring Framework 2 Inference Engine 2 NN Authoring Framework 3 Inference Engine 3 Every Tool Needs an Exporter to Every Accelerator Neural Network Inferencing Fragmentation toll on Applications Inference Engine 1 Hardware 1 Vision/AI Inference Engine 2 Hardware 2 Application Inference Engine 3 Hardware 3 Every Application Needs know about Every Accelerator API © Copyright Khronos Group 2017 - Page 4 Khronos APIs Connect Software to Silicon Software Silicon Khronos is an International Industry Consortium of over 100 companies creating royalty-free, open standard APIs to enable software to access
    [Show full text]
  • Khronos Template 2015
    Ecosystem Overview Neil Trevett | Khronos President NVIDIA Vice President Developer Ecosystem [email protected] | @neilt3d © Copyright Khronos Group 2016 - Page 1 Khronos Mission Software Silicon Khronos is an Industry Consortium of over 100 companies creating royalty-free, open standard APIs to enable software to access hardware acceleration for graphics, parallel compute and vision © Copyright Khronos Group 2016 - Page 2 http://accelerateyourworld.org/ © Copyright Khronos Group 2016 - Page 3 Vision Pipeline Challenges and Opportunities Growing Camera Diversity Diverse Vision Processors Sensor Proliferation 22 Flexible sensor and camera Use efficient acceleration to Combine vision output control to GENERATE PROCESS with other sensor data an image stream the image stream on device © Copyright Khronos Group 2016 - Page 4 OpenVX – Low Power Vision Acceleration • Higher level abstraction API - Targeted at real-time mobile and embedded platforms • Performance portability across diverse architectures - Multi-core CPUs, GPUs, DSPs and DSP arrays, ISPs, Dedicated hardware… • Extends portable vision acceleration to very low power domains - Doesn’t require high-power CPU/GPU Complex - Lower precision requirements than OpenCL - Low-power host can setup and manage frame-rate graph Vision Engine Middleware Application X100 Dedicated Vision Processing Hardware Efficiency Vision DSPs X10 GPU Compute Accelerator Multi-core Accelerator Power Efficiency Power X1 CPU Accelerator Computation Flexibility © Copyright Khronos Group 2016 - Page 5 OpenVX Graphs
    [Show full text]
  • Migrating from Opengl to Vulkan Mark Kilgard, January 19, 2016 About the Speaker Who Is This Guy?
    Migrating from OpenGL to Vulkan Mark Kilgard, January 19, 2016 About the Speaker Who is this guy? Mark Kilgard Principal Graphics Software Engineer in Austin, Texas Long-time OpenGL driver developer at NVIDIA Author and implementer of many OpenGL extensions Collaborated on the development of Cg First commercial GPU shading language Recently working on GPU-accelerated vector graphics (Yes, and wrote GLUT in ages past) 2 Motivation for Talk Coming from OpenGL, Preparing for Vulkan What kinds of apps benefit from Vulkan? How to prepare your OpenGL code base to transition to Vulkan How various common OpenGL usage scenarios are re-thought in Vulkan Re-thinking your application structure for Vulkan 3 Analogy Different Valid Approaches 4 Analogy Fixed-function OpenGL Pre-assembled toy car fun out of the box, not much room for customization 5 AZDO = Approaching Zero Driver Overhead Analogy Modern AZDO OpenGL with Programmable Shaders LEGO Kit you build it yourself, comes with plenty of useful, pre-shaped pieces 6 Analogy Vulkan Pine Wood Derby Kit you build it yourself to race from raw materials power tools used to assemble, adult supervision highly recommended 7 Analogy Different Valid Approaches Fixed-function OpenGL Modern AZDO OpenGL with Vulkan Programmable Shaders 8 Beneficial Vulkan Scenarios Has Parallelizable CPU-bound Graphics Work yes Can your graphics Is your graphics work start work creation be CPU bound? parallelized? yes Vulkan friendly 9 Beneficial Vulkan Scenarios Maximizing a Graphics Platform Budget You’ll yes do whatever Your graphics start it takes to squeeze platform is fixed out max perf. yes Vulkan friendly 10 Beneficial Vulkan Scenarios Managing Predictable Performance, Free of Hitching You put yes You can a premium on manage your start avoiding graphics resource hitches allocations yes Vulkan friendly 11 Unlikely to Benefit Scenarios to Reconsider Coding to Vulkan 1.
    [Show full text]
  • Building the Metaverse One Open Standard at a Time
    Building the Metaverse One Open Standard at a Time Khronos APIs and 3D Asset Formats for XR Neil Trevett Khronos President NVIDIA VP Developer Ecosystems [email protected]| @neilt3d April 2020 © Khronos® Group 2020 This work is licensed under a Creative Commons Attribution 4.0 International License © The Khronos® Group Inc. 2020 - Page 1 Khronos Connects Software to Silicon Open interoperability standards to enable software to effectively harness the power of multiprocessors and accelerator silicon 3D graphics, XR, parallel programming, vision acceleration and machine learning Non-profit, member-driven standards-defining industry consortium Open to any interested company All Khronos standards are royalty-free Well-defined IP Framework protects participant’s intellectual property >150 Members ~ 40% US, 30% Europe, 30% Asia This work is licensed under a Creative Commons Attribution 4.0 International License © The Khronos® Group Inc. 2020 - Page 2 Khronos Active Initiatives 3D Graphics 3D Assets Portable XR Parallel Computation Desktop, Mobile, Web Authoring Augmented and Vision, Inferencing, Machine Embedded and Safety Critical and Delivery Virtual Reality Learning Guidelines for creating APIs to streamline system safety certification This work is licensed under a Creative Commons Attribution 4.0 International License © The Khronos® Group Inc. 2020 - Page 3 Pervasive Vulkan Desktop and Mobile GPUs http://vulkan.gpuinfo.org/ Platforms Apple Desktop Android (via porting Media Players Consoles (Android 7.0+) layers) (Vulkan 1.1 required on Android Q) Virtual Reality Cloud Services Game Streaming Embedded Engines Croteam Serious Engine Note: The version of Vulkan available will depend on platform and vendor This work is licensed under a Creative Commons Attribution 4.0 International License © The Khronos® Group Inc.
    [Show full text]
  • The Opengl ES Shading Language
    The OpenGL ES® Shading Language Language Version: 3.20 Document Revision: 12 246 JuneAugust 2015 Editor: Robert J. Simpson, Qualcomm OpenGL GLSL editor: John Kessenich, LunarG GLSL version 1.1 Authors: John Kessenich, Dave Baldwin, Randi Rost 1 Copyright (c) 2013-2015 The Khronos Group Inc. All Rights Reserved. This specification is protected by copyright laws and contains material proprietary to the Khronos Group, Inc. It or any components may not be reproduced, republished, distributed, transmitted, displayed, broadcast, or otherwise exploited in any manner without the express prior written permission of Khronos Group. You may use this specification for implementing the functionality therein, without altering or removing any trademark, copyright or other notice from the specification, but the receipt or possession of this specification does not convey any rights to reproduce, disclose, or distribute its contents, or to manufacture, use, or sell anything that it may describe, in whole or in part. Khronos Group grants express permission to any current Promoter, Contributor or Adopter member of Khronos to copy and redistribute UNMODIFIED versions of this specification in any fashion, provided that NO CHARGE is made for the specification and the latest available update of the specification for any version of the API is used whenever possible. Such distributed specification may be reformatted AS LONG AS the contents of the specification are not changed in any way. The specification may be incorporated into a product that is sold as long as such product includes significant independent work developed by the seller. A link to the current version of this specification on the Khronos Group website should be included whenever possible with specification distributions.
    [Show full text]
  • The State of Vulkan on Apple 03June 2021
    The State of Vulkan on Apple Devices Why you should be developing for the Apple ecosystem with Vulkan, and how to do it. Richard S. Wright Jr., LunarG June 3, 2021 If you are reading this, you likely already know what Vulkan is -- a low-overhead, high-performance, cross-platform, graphics and compute API designed to take maximum advantage of today's highly parallel and scalable graphics and compute hardware. To be sure, there are other APIs with this goal such as Microsoft’s DirectX and Apple’s Metal API. For a long time OpenGL matured to meet the needs of rapidly evolving hardware and met well the challenges of cross-platform development of high-performance graphics applications. OpenGL will remain viable for a very long time for many applications (and can actually be layered on top of Vulkan), but to take maximum advantage of modern hardware, a new and somewhat reinvented approach is called for. Everyone does not need the level of control that Vulkan provides, but if you do, you can think of Vulkan as a lower-level successor to OpenGL in that it attempts to enable highly portable, yet high-performance, application development across desktop and mobile device platforms. Writing platform specific code does have its advantages, and vendors such as Microsoft and Apple certainly adore having best of breed titles available exclusively for their customers as it adds value to their ecosystem. For developers though, it is often in our best interests to make our products available as widely as possible. Aside from gaming consoles with their mostly closed development environments, Windows, Linux, Apple, and Android combined represent most of the market for developers of end user software and games today.
    [Show full text]
  • News Release for More Information: Neil Trevett, President, Khronos | [email protected] | Phone: +1 (408) 464 7053
    News Release For more information: Neil Trevett, President, Khronos | [email protected] | Phone: +1 (408) 464 7053 Khronos Releases OpenVX 1.1 Specification for High Performance, Low Power Computer Vision Acceleration Expanded range of processing functions; Enhanced flexibility for data access and processing; Full conformance tests available; Safety Critical specification in development May 2nd 2016 – Embedded Vision Summit, Santa Clara, CA – The Khronos™ Group, an open consortium of leading hardware and software companies, announces the immediate availability of the OpenVX™ 1.1 specification for cross platform acceleration of computer vision applications and libraries. OpenVX enables performance and power optimized computer vision algorithms for use cases such as face, body and gesture tracking, smart video surveillance, automatic driver assistance systems, object and scene reconstruction, augmented reality, visual inspection, robotics and more. Conformant OpenVX 1.0 implementations and tools are shipping from AMD, Imagination, Intel, NVIDIA, Synopsis and VeriSilicon. OpenVX 1.1 builds on this momentum by adding new processing functions for use cases such as computational photography, and enhances application control over how data is accessed and processed. An open source OpenVX 1.1 sample implementation and full conformance tests will be available before mid-2016. Details on the OpenVX specifications and Adopters Program are available at: www.khronos.org/openvx. “More and more products are incorporating computer vision, and OpenVX
    [Show full text]
  • DES-N1740 I.MX 8 and the Vulkan
    i.MX 8 AND THE VULKAN API THE FUTURE OF HIGH PERFORMANCE GRAPHICS IMPLEMENTATIONS GABRIEL DAGANI SENIOR GRAPHICS ENGINEER / SOC ARCHITECT FTF-DES-N1740 MAY 19, 2016 PUBLIC USE AGENDA • What is Vulkan? • Vulkan vs. OpenGL ES and when to use which API • Using Vulkan with i.MX 8 1 PUBLIC USE FTF-DES-N1740 1 PUBLIC USE FTF-DES-N1740 What Is Vulkan? • Vulkan is … − Cross platform API − Open Source Standard − Meant for Graphics & Compute Workloads • Vulkan is NOT… − The successor of OpenGL − The race of Spock (‘k’ not ‘c’) 2 PUBLIC USE FTF-DES-N1740 Next Generation GPU APIs 3 PUBLIC USE FTF-DES-N1740 Source: Khronos Group How Did Vulkan Come to Be? • As graphics APIs evolved, backwards compatibility as a paradigm became unwieldy. • Embedded Version of OGL emerged to simplify and enhance usage for size and constraints. OpenGL 4.4 OpenGL 4.0 OpenGL 3.2 OGL ES 3.2 OpenGL 3.0 OGLES 3.1 OpenGL 2.1 OGLES 3.0 Vulkan OpenGL 2.0 OGL ES 2.0 1.0 OpenGL 1.5 OGLES 1.1 OpenGL 1.1 OGLES 1.0 4 PUBLIC USE FTF-DES-N1740 Excess Baggage from Backwards Compatibility • Supporting general scenarios instead of optimizing for specific scenarios − Driver overhead for unnecessary driver paths • Resource management based on generic machine type − Prebaked garbage collection, cmd buffer management, multithreading, scheduling, etc. − All Modes: Desktop / Embedded / Compute / Graphics • Lacking support for multiple GPU cores in a context • Increased overhead for all use-cases to ensure compatibility − Implies decreased performance and increased power consumption 5 PUBLIC
    [Show full text]
  • Khronos API Landscape EVE Oct19
    APIs for Accelerating Vision and Inferencing An Overview of Options and Trade-offs Neil Trevett Khronos President NVIDIA VP Developer Ecosystems [email protected] | @neilt3d This work is licensed under a Creative Commons Attribution 4.0 International License © The Khronos® Group Inc. 2019 - Page 1 Khronos is an open, non-profit, member-driven industry consortium developing royalty-free standards, and vibrant ecosystems, to harness the power of silicon acceleration for demanding graphics rendering and computationally intensive applications such as inferencing and vision processing >150 Members ~ 40% US, 30% Europe, 30% Asia Some Khronos Standards Relevant to Embedded Vision and Inferencing Vision and Neural Network Heterogeneous Parallel C++ Parallel GPU Inferencing Runtime Exchange Format Parallel Compute Compute IR Compute Acceleration This work is licensed under a Creative Commons Attribution 4.0 International License © The Khronos® Group Inc. 2019 - Page 2 Embedded Vision and Inferencing Acceleration Networks trained on Neural Network high-end desktop Training Data and cloud systems Training Trained Networks Compilation Ingestion Apps link to compiled code or Compiled Inferencing Application Vision inferencing library Code Engine Code Library Hardware Acceleration APIs Sensor Data Diverse Embedded DSP Hardware Dedicated (GPUs, DSPs, FPGAs) FPGA Hardware GPU This work is licensed under a Creative Commons Attribution 4.0 International License © The Khronos® Group Inc. 2019 - Page 3 Neural Network Training Acceleration Neural Network Training Frameworks Neural Net Training Network training needs 32- Neural Network Neural Net Training bit floating-point precision NeuralNeural NetFrameworks Net Training Training Authoring Interchange Frameworks Training Data FrameworksFrameworks Neural Network cuDNN MIOpen clDNN Acceleration Libraries Desktop and Cloud Hardware Hardware Acceleration APIs TPU This work is licensed under a Creative Commons Attribution 4.0 International License © The Khronos® Group Inc.
    [Show full text]
  • Compositing and the Reuse of Software in Safety Critical Graphics Applications
    Compositing and the Reuse of Software in Safety Critical Graphics Applications INTRODUCTION Avionics and automotive hardware is ever-evolving, but often the accompanying display software doesn’t need to change. CoreAVI has a long history of collaboration with the Future Airborne Capability Environment (FACE) Consorti- um to lead the way in the creation of truly portable display software. Open standards including OpenGL®, FACE™, and Vulkan® allow a clear separation between display software and the hardware on which it runs. This separation benefits software developers as it allows them to start development work without needing to know what GPU they will be writing code for. CoreAVI has introduced a way to use the GPU to composite the visual output of different pieces of display software without requiring them to be rewritten each time a new symbol is added to the display. This method of compositing maintains the transparency of each pixel and sends them to the compositing application for proper blending, which in turn enables full screen overlays to be run as separate applications, as well as the splitting of the screen’s geometry. Such composition allows all but the compositing application to remain unchanged as new technologies are added or parts of the display rearranged. This white paper details how existing safety critical DO-178C or ISO 26262 application software source code can effectively be rehosted on advancing hardware. WHAT IS THE EGL COMPOSITOR? On February 3, 2017, the FACE Consortium along with the Khronos Group published the EGL_EXT_compositor exten- sion to add compositing capability to EGL. This extension minimizes application effort and allows for the composition of multiple windows within a multi-partition EGL system.
    [Show full text]
  • Advanced Graphics Opengl and Shaders I
    Advanced Graphics OpenGL and Shaders I 1 Alex Benton, University of Cambridge – [email protected] Supported in part by Google UK, Ltd 3D technologies today Java OpenGL ● Common, re-usable language; ● Open source with many well-designed implementations ● Steadily increasing popularity in ● Well-designed, old, and still evolving industry ● Fairly cross-platform ● Weak but evolving 3D support DirectX/Direct3d (Microsoft) C++ ● Microsoft™ only ● Long-established language ● Dependable updates ● Long history with OpenGL Mantle (AMD) ● Long history with DirectX ● Targeted at game developers ● Losing popularity in some fields (finance, web) but still strong in ● AMD-specific others (games, medical) Higher-level commercial libraries JavaScript ● RenderMan ● WebGL is surprisingly popular ● AutoDesk / SoftImage 2 OpenGL OpenGL is… A state-based renderer ● Hardware-independent ● many settings are configured ● Operating system independent before passing in data; rendering ● Vendor neutral behavior is modified by existing On many platforms state ● Great support on Windows, Mac, Accelerates common 3D graphics linux, etc operations ● Support for mobile devices with ● Clipping (for primitives) OpenGL ES ● Hidden-surface removal ● Android, iOS (but not (Z-buffering) Windows Phone) ● Texturing, alpha blending ● Android Wear watches! NURBS and other advanced ● Web support with WebGL primitives (GLUT) 3 Mobile GPUs ● OpenGL ES 1.0-3.2 ● A stripped-down version of OpenGL ● Removes functionality that is not strictly necessary on mobile devices (like recursion!) ● Devices ● iOS: iPad, iPhone, iPod Touch OpenGL ES 2.0 rendering (iOS) ● Android phones ● PlayStation 3, Nintendo 3DS, and more 4 WebGL ● JavaScript library for 3D rendering in a web browser ● Based on OpenGL ES 2.0 ● Many supporting JS libraries ● Even gwt, angular, dart..
    [Show full text]