Tulipp_H2020_Hipeac'17 Conference_PEPGUM Workshop_January 017

T
This project has received funding from
the European Union’s Horizon 20 20
research and innovation programme
under grant agreement No 688403
www.tulipp.eu
TULIPP
Place :
Date :
PEGPUM Workshop
https://www.hipeac.net/events/activities/7435/pegpum/
Project Overview
HiPEAC’17, Stockholm
24th of January 2017
Philippe Millet, Thales, Coordinator
8 Partners
• Thales : Coordinator and Medical
use case
• Sundance : Hardware
• Hipperos : Operating system
• Synective Labs : ADAS use case
• Efficient Innovation :
Management
• Fraunhofer IOSB : UAV use case
• Ruhr Universität Bochum : FPGA
tools
• NTNU : Performance tools
Outline
• Project objectives
• Main outcomes
• Advisory Board and Ecosystem
WP presentation
WP7: Management, Coordination
LABEL : Marketing, Ecosystem and Pre-normalisation
WP6: IP protection, Dissemination, Communication, Advisory Board
and Exploitation preparation
WP1: Reference platform definition
(Interfaces & implementation Rules)
Instantiations
WP2:
Hardware
WP4:
Programming
Toolchain
WP3:
Runtime, API,
Libraries & OS
feedback WP5 : Usecases description
and Integration and platform
validation
Starter Kit
Handbook Overview
Guidelines
Guidelines
Advice: Exploit both vectorization and multithreading for high performance on multicore
processors with vector units such as the ARM Cortex A9. On these architectures, utilizing
all hardware execution resources is key to achieve high performance [2] [4, 5].
Recommended implementation method: Use OpenMP. OpenMP is a widely supported
parallel programming API that enables programmers to express vectorization and
multithreading operations concisely using compiler directives. Programmers need not
worry about specifying scheduling and synchronization operations in code. These are
handled transparently by the OpenMP runtime system. See the official OpenMP
examples[6] to understand in more detail about exploiting vectorization and
multithreading simultaneously.
Reference Platform
Towards Ubiquitous Low-Power Image Processing Platforms
Component tools
Operating System
Processor
Toolchain
Reference Platform
Memory
IO
Processor
• The use case is taken from the many
applications that now enter our cars, doing
emergency braking, lane keeping, etc., with
direction towards autonomous cars
• The chosen image processing algorithm for the
use case is pedestrian detection that typically is
used for emergency braking and driver
assistance systems.
• Pedestrian detection is today mostly made by
either using Viola/Jones classifiers or Deep
Learning
• Viola Jones classifying has been chosen due to
its more challenging memory access patterns
ADAS use case
Pedestrian
detection
Safety
application
Car
integration
The Use Case
ADAS use case
Medical imaging use case
• Real-Time X-Ray imaging for surgery
• Reduce radiation dose by 75%
• Add noise removal processing with critical
real-time constraints
Medical imaging use case
TDLP
RAW IMAGE
THALES Processing
Unit
CI / ICS
UI
GigE-Vision + Msg
THALES Flat panel detector
Customer system
UI
GigE-Vision + Msg
CI / ICS
Nano Processing Unit
Inside the detector
Based on SoC (credit card size board)
Customer system
THALES Flat panel detector
Before Tulipp
After Tulipp
Unmanned Aerial Vehicle (UAV) use case
• Performs real-time stereo depth estimation to do obstacle /
collision avoidance (for an UAV), i.e. to detect obstacles in
direction of flight
• Based on dual cameras
Unmanned Aerial Vehicle (UAV) use case
• Uses state-of-the art stereo algorithms (image correlation)
• Produces a distance image, i.e. where the image data
shows the distance to each object
TULIPP Hardware Node
• Now – Zynq SoC SoM • Soon– Zynq MPSoC SoM
40mm x 50 mm – Z7030 40mm x 50 mm - ZU5EV
TULIPP System Node
Low-Power Image Processing RTOS
Needs OS
- high reliability,
- low power,
- hard real-time
- high performance
This kind of RTOS is not (yet) available today, as current GPOS (e.g.
Linux) or RTOS lack one or more of the required features and
performance.
Specific Image processing Needs
- supporting the hardware accelerators
- the libraries needed for image processing.
STHEM: The TULIPP Tool-chain
Status:
• Xilinx SDSoC has been extended to
support the current platform
• Support for HIPPEROS OS is underway
Insights:
• Significant effort has been invested
into the development of vendor tools
• STHEM fills the productivity gaps
between existing tools
Support uTilities for Heterogeneous
Embedded image processing (STHEM) •Supports development for all platform
components
•Map source files of the application to the
appropriate tool chain
•Retrieve OS configuration from the developer
Development and Mapping
•Boot OS with selected configuration (if
needed due to changed configuration)
•Update files (binaries, bitfiles, etc.)
•Initialise the reconfigurable logic (if needed)
•Start the application with the requested
instrumentation
Runner
•Analyses performance results and presents
findings to the developer
Analyser
TULIPP in the news
20
Advisory Board and EcoSystem
Advisory
Board
(WP6)
Reference
Platform
(WP1)
Tulipp Guide, implementation and demos
Ask for review / advise
Roles in the project:
Provide information about standards
Give feedback on the approach
Early adopters
Sys temEco
AB: first step to
the ecosystem
Advisory Board Members
22
Advisory Board and EcoSystem
• Pledge
• Interconnectivity
• Faster time-to-market
• Open standards
1 sur 23

Recommandé

Samos July 2016_tulipp-H2020 project presentation par
Samos July 2016_tulipp-H2020 project presentationSamos July 2016_tulipp-H2020 project presentation
Samos July 2016_tulipp-H2020 project presentationTulipp. Eu
569 vues13 diapositives
Tulipp collaboration Workshop - Advanced Computing and CPS - June 2016 par
Tulipp collaboration Workshop - Advanced Computing and CPS - June 2016Tulipp collaboration Workshop - Advanced Computing and CPS - June 2016
Tulipp collaboration Workshop - Advanced Computing and CPS - June 2016 Tulipp. Eu
478 vues5 diapositives
Phidias: Steps forward in detection and identification of anomalous atmospher... par
Phidias: Steps forward in detection and identification of anomalous atmospher...Phidias: Steps forward in detection and identification of anomalous atmospher...
Phidias: Steps forward in detection and identification of anomalous atmospher...Phidias
317 vues42 diapositives
Science Demonstrator Session: Physics and Astrophysics par
Science Demonstrator Session: Physics and AstrophysicsScience Demonstrator Session: Physics and Astrophysics
Science Demonstrator Session: Physics and AstrophysicsEOSCpilot .eu
105 vues31 diapositives
Experience in managing service portfolio by Pasquale Pagano par
Experience in managing service portfolio by Pasquale PaganoExperience in managing service portfolio by Pasquale Pagano
Experience in managing service portfolio by Pasquale PaganoBlue BRIDGE
236 vues11 diapositives
Machinel Learning with spark par
Machinel Learning with spark Machinel Learning with spark
Machinel Learning with spark Ons Dridi
807 vues53 diapositives

Contenu connexe

Similaire à Tulipp_H2020_Hipeac'17 Conference_PEPGUM Workshop_January 017

TULIPP H2020 Project presentation @ FPGA Network: Implementing Machine Vision... par
TULIPP H2020 Project presentation @ FPGA Network: Implementing Machine Vision...TULIPP H2020 Project presentation @ FPGA Network: Implementing Machine Vision...
TULIPP H2020 Project presentation @ FPGA Network: Implementing Machine Vision...Tulipp. Eu
396 vues20 diapositives
TULIPP at NMI 18-5-17 par
TULIPP at NMI 18-5-17TULIPP at NMI 18-5-17
TULIPP at NMI 18-5-17Sundance Multiprocessor Technology Ltd.
266 vues20 diapositives
Sundance TULIPP Workshop at Nottingham Trent University par
Sundance TULIPP Workshop at Nottingham Trent UniversitySundance TULIPP Workshop at Nottingham Trent University
Sundance TULIPP Workshop at Nottingham Trent UniversitySundance Multiprocessor Technology Ltd.
402 vues18 diapositives
TULIPP at the 10th Intelligent Imaging Event par
TULIPP at the 10th Intelligent Imaging EventTULIPP at the 10th Intelligent Imaging Event
TULIPP at the 10th Intelligent Imaging EventSundance Multiprocessor Technology Ltd.
187 vues43 diapositives
HIPPEROS's at EMVA 2017 par
HIPPEROS's at EMVA 2017 HIPPEROS's at EMVA 2017
HIPPEROS's at EMVA 2017 Tulipp. Eu
121 vues30 diapositives
HiPEAC 2019 Workshop Overview par
HiPEAC 2019 Workshop OverviewHiPEAC 2019 Workshop Overview
HiPEAC 2019 Workshop OverviewTulipp. Eu
66 vues44 diapositives

Similaire à Tulipp_H2020_Hipeac'17 Conference_PEPGUM Workshop_January 017(20)

TULIPP H2020 Project presentation @ FPGA Network: Implementing Machine Vision... par Tulipp. Eu
TULIPP H2020 Project presentation @ FPGA Network: Implementing Machine Vision...TULIPP H2020 Project presentation @ FPGA Network: Implementing Machine Vision...
TULIPP H2020 Project presentation @ FPGA Network: Implementing Machine Vision...
Tulipp. Eu396 vues
HIPPEROS's at EMVA 2017 par Tulipp. Eu
HIPPEROS's at EMVA 2017 HIPPEROS's at EMVA 2017
HIPPEROS's at EMVA 2017
Tulipp. Eu121 vues
HiPEAC 2019 Workshop Overview par Tulipp. Eu
HiPEAC 2019 Workshop OverviewHiPEAC 2019 Workshop Overview
HiPEAC 2019 Workshop Overview
Tulipp. Eu66 vues
TULIPP H2020 Project: Low power high performance real-time computer vision on... par Tulipp. Eu
TULIPP H2020 Project: Low power high performance real-time computer vision on...TULIPP H2020 Project: Low power high performance real-time computer vision on...
TULIPP H2020 Project: Low power high performance real-time computer vision on...
Tulipp. Eu403 vues
Developing Tizen OS Based Solutions (IDF13) - Chris Norman par Ryo Jin
Developing Tizen OS Based Solutions (IDF13) - Chris NormanDeveloping Tizen OS Based Solutions (IDF13) - Chris Norman
Developing Tizen OS Based Solutions (IDF13) - Chris Norman
Ryo Jin2.4K vues
SCAPE - Scalable Preservation Environments par SCAPE Project
SCAPE - Scalable Preservation EnvironmentsSCAPE - Scalable Preservation Environments
SCAPE - Scalable Preservation Environments
SCAPE Project483 vues
2017-05-10 Gate4SPICE: "Legacy Software" par Alexander Much
2017-05-10 Gate4SPICE: "Legacy Software"2017-05-10 Gate4SPICE: "Legacy Software"
2017-05-10 Gate4SPICE: "Legacy Software"
Alexander Much325 vues
Case study: How Cozy Cloud monitors every layer of its activity using OVH Met... par OVHcloud
Case study: How Cozy Cloud monitors every layer of its activity using OVH Met...Case study: How Cozy Cloud monitors every layer of its activity using OVH Met...
Case study: How Cozy Cloud monitors every layer of its activity using OVH Met...
OVHcloud302 vues
Виктор Ерухимов Open VX mixar moscow sept'15 par mixARConference
Виктор Ерухимов Open VX  mixar moscow sept'15 Виктор Ерухимов Open VX  mixar moscow sept'15
Виктор Ерухимов Open VX mixar moscow sept'15
mixARConference602 vues
e-Infrastructure available for research, using the right tool for the right job par David Wallom
e-Infrastructure available for research, using the right tool for the right jobe-Infrastructure available for research, using the right tool for the right job
e-Infrastructure available for research, using the right tool for the right job
David Wallom283 vues
Unbundling the Modern Streaming Stack With Dunith Dhanushka | Current 2022 par HostedbyConfluent
Unbundling the Modern Streaming Stack With Dunith Dhanushka | Current 2022Unbundling the Modern Streaming Stack With Dunith Dhanushka | Current 2022
Unbundling the Modern Streaming Stack With Dunith Dhanushka | Current 2022

Plus de Tulipp. Eu

What are TULIPP starter kits? par
What are TULIPP starter kits?What are TULIPP starter kits?
What are TULIPP starter kits?Tulipp. Eu
314 vues24 diapositives
Quantifying Energy Consumption for Practical Fork-Join Parallelism on an Embe... par
Quantifying Energy Consumption for Practical Fork-Join Parallelism on an Embe...Quantifying Energy Consumption for Practical Fork-Join Parallelism on an Embe...
Quantifying Energy Consumption for Practical Fork-Join Parallelism on an Embe...Tulipp. Eu
226 vues132 diapositives
TULIPP overview par
TULIPP overviewTULIPP overview
TULIPP overviewTulipp. Eu
119 vues19 diapositives
HiPEAC 2019 Workshop - Vision Processing par
HiPEAC 2019 Workshop - Vision ProcessingHiPEAC 2019 Workshop - Vision Processing
HiPEAC 2019 Workshop - Vision ProcessingTulipp. Eu
297 vues24 diapositives
HiPEAC 2019 Workshop - Real-Time Modelling Visual Scenes with Biological Insp... par
HiPEAC 2019 Workshop - Real-Time Modelling Visual Scenes with Biological Insp...HiPEAC 2019 Workshop - Real-Time Modelling Visual Scenes with Biological Insp...
HiPEAC 2019 Workshop - Real-Time Modelling Visual Scenes with Biological Insp...Tulipp. Eu
79 vues22 diapositives
HiPEAC 2019 Tutorial - Maestro RTOS par
HiPEAC 2019 Tutorial - Maestro RTOSHiPEAC 2019 Tutorial - Maestro RTOS
HiPEAC 2019 Tutorial - Maestro RTOSTulipp. Eu
127 vues59 diapositives

Plus de Tulipp. Eu(12)

What are TULIPP starter kits? par Tulipp. Eu
What are TULIPP starter kits?What are TULIPP starter kits?
What are TULIPP starter kits?
Tulipp. Eu314 vues
Quantifying Energy Consumption for Practical Fork-Join Parallelism on an Embe... par Tulipp. Eu
Quantifying Energy Consumption for Practical Fork-Join Parallelism on an Embe...Quantifying Energy Consumption for Practical Fork-Join Parallelism on an Embe...
Quantifying Energy Consumption for Practical Fork-Join Parallelism on an Embe...
Tulipp. Eu226 vues
HiPEAC 2019 Workshop - Vision Processing par Tulipp. Eu
HiPEAC 2019 Workshop - Vision ProcessingHiPEAC 2019 Workshop - Vision Processing
HiPEAC 2019 Workshop - Vision Processing
Tulipp. Eu297 vues
HiPEAC 2019 Workshop - Real-Time Modelling Visual Scenes with Biological Insp... par Tulipp. Eu
HiPEAC 2019 Workshop - Real-Time Modelling Visual Scenes with Biological Insp...HiPEAC 2019 Workshop - Real-Time Modelling Visual Scenes with Biological Insp...
HiPEAC 2019 Workshop - Real-Time Modelling Visual Scenes with Biological Insp...
Tulipp. Eu79 vues
HiPEAC 2019 Tutorial - Maestro RTOS par Tulipp. Eu
HiPEAC 2019 Tutorial - Maestro RTOSHiPEAC 2019 Tutorial - Maestro RTOS
HiPEAC 2019 Tutorial - Maestro RTOS
Tulipp. Eu127 vues
HiPEAC 2019 Tutorial - Image Processing Library:HiFlipVX par Tulipp. Eu
HiPEAC 2019 Tutorial - Image Processing Library:HiFlipVXHiPEAC 2019 Tutorial - Image Processing Library:HiFlipVX
HiPEAC 2019 Tutorial - Image Processing Library:HiFlipVX
Tulipp. Eu138 vues
HiPEAC 2019 Tutorial - Sthem overview par Tulipp. Eu
HiPEAC 2019 Tutorial - Sthem overviewHiPEAC 2019 Tutorial - Sthem overview
HiPEAC 2019 Tutorial - Sthem overview
Tulipp. Eu83 vues
HiPEAC 2019 Workshop - Hardware Starter Kit Agri par Tulipp. Eu
HiPEAC 2019 Workshop - Hardware Starter Kit Agri HiPEAC 2019 Workshop - Hardware Starter Kit Agri
HiPEAC 2019 Workshop - Hardware Starter Kit Agri
Tulipp. Eu100 vues
HiPEAC 2019 Workshop - Use Cases par Tulipp. Eu
HiPEAC 2019 Workshop - Use CasesHiPEAC 2019 Workshop - Use Cases
HiPEAC 2019 Workshop - Use Cases
Tulipp. Eu52 vues
Tulipp starter-kit-agri par Tulipp. Eu
Tulipp starter-kit-agriTulipp starter-kit-agri
Tulipp starter-kit-agri
Tulipp. Eu560 vues
D1.1 reference platform_v1_20161215 par Tulipp. Eu
D1.1 reference platform_v1_20161215D1.1 reference platform_v1_20161215
D1.1 reference platform_v1_20161215
Tulipp. Eu157 vues

Tulipp_H2020_Hipeac'17 Conference_PEPGUM Workshop_January 017

  • 1. This project has received funding from the European Union’s Horizon 20 20 research and innovation programme under grant agreement No 688403 www.tulipp.eu TULIPP Place : Date : PEGPUM Workshop https://www.hipeac.net/events/activities/7435/pegpum/ Project Overview HiPEAC’17, Stockholm 24th of January 2017 Philippe Millet, Thales, Coordinator
  • 2. 8 Partners • Thales : Coordinator and Medical use case • Sundance : Hardware • Hipperos : Operating system • Synective Labs : ADAS use case • Efficient Innovation : Management • Fraunhofer IOSB : UAV use case • Ruhr Universität Bochum : FPGA tools • NTNU : Performance tools
  • 3. Outline • Project objectives • Main outcomes • Advisory Board and Ecosystem
  • 4. WP presentation WP7: Management, Coordination LABEL : Marketing, Ecosystem and Pre-normalisation WP6: IP protection, Dissemination, Communication, Advisory Board and Exploitation preparation WP1: Reference platform definition (Interfaces & implementation Rules) Instantiations WP2: Hardware WP4: Programming Toolchain WP3: Runtime, API, Libraries & OS feedback WP5 : Usecases description and Integration and platform validation
  • 8. Guidelines Advice: Exploit both vectorization and multithreading for high performance on multicore processors with vector units such as the ARM Cortex A9. On these architectures, utilizing all hardware execution resources is key to achieve high performance [2] [4, 5]. Recommended implementation method: Use OpenMP. OpenMP is a widely supported parallel programming API that enables programmers to express vectorization and multithreading operations concisely using compiler directives. Programmers need not worry about specifying scheduling and synchronization operations in code. These are handled transparently by the OpenMP runtime system. See the official OpenMP examples[6] to understand in more detail about exploiting vectorization and multithreading simultaneously.
  • 9. Reference Platform Towards Ubiquitous Low-Power Image Processing Platforms Component tools Operating System Processor Toolchain Reference Platform Memory IO Processor
  • 10. • The use case is taken from the many applications that now enter our cars, doing emergency braking, lane keeping, etc., with direction towards autonomous cars • The chosen image processing algorithm for the use case is pedestrian detection that typically is used for emergency braking and driver assistance systems. • Pedestrian detection is today mostly made by either using Viola/Jones classifiers or Deep Learning • Viola Jones classifying has been chosen due to its more challenging memory access patterns ADAS use case
  • 12. Medical imaging use case • Real-Time X-Ray imaging for surgery • Reduce radiation dose by 75% • Add noise removal processing with critical real-time constraints
  • 13. Medical imaging use case TDLP RAW IMAGE THALES Processing Unit CI / ICS UI GigE-Vision + Msg THALES Flat panel detector Customer system UI GigE-Vision + Msg CI / ICS Nano Processing Unit Inside the detector Based on SoC (credit card size board) Customer system THALES Flat panel detector Before Tulipp After Tulipp
  • 14. Unmanned Aerial Vehicle (UAV) use case • Performs real-time stereo depth estimation to do obstacle / collision avoidance (for an UAV), i.e. to detect obstacles in direction of flight • Based on dual cameras
  • 15. Unmanned Aerial Vehicle (UAV) use case • Uses state-of-the art stereo algorithms (image correlation) • Produces a distance image, i.e. where the image data shows the distance to each object
  • 16. TULIPP Hardware Node • Now – Zynq SoC SoM • Soon– Zynq MPSoC SoM 40mm x 50 mm – Z7030 40mm x 50 mm - ZU5EV
  • 18. Low-Power Image Processing RTOS Needs OS - high reliability, - low power, - hard real-time - high performance This kind of RTOS is not (yet) available today, as current GPOS (e.g. Linux) or RTOS lack one or more of the required features and performance. Specific Image processing Needs - supporting the hardware accelerators - the libraries needed for image processing.
  • 19. STHEM: The TULIPP Tool-chain Status: • Xilinx SDSoC has been extended to support the current platform • Support for HIPPEROS OS is underway Insights: • Significant effort has been invested into the development of vendor tools • STHEM fills the productivity gaps between existing tools Support uTilities for Heterogeneous Embedded image processing (STHEM) •Supports development for all platform components •Map source files of the application to the appropriate tool chain •Retrieve OS configuration from the developer Development and Mapping •Boot OS with selected configuration (if needed due to changed configuration) •Update files (binaries, bitfiles, etc.) •Initialise the reconfigurable logic (if needed) •Start the application with the requested instrumentation Runner •Analyses performance results and presents findings to the developer Analyser
  • 20. TULIPP in the news 20
  • 21. Advisory Board and EcoSystem Advisory Board (WP6) Reference Platform (WP1) Tulipp Guide, implementation and demos Ask for review / advise Roles in the project: Provide information about standards Give feedback on the approach Early adopters Sys temEco AB: first step to the ecosystem
  • 23. Advisory Board and EcoSystem • Pledge • Interconnectivity • Faster time-to-market • Open standards