SlideShare une entreprise Scribd logo
1  sur  11
Bringing Intelligent Vision Processing to
Low-Power Embedded Systems February 2015
2CEVA Proprietary Information
Introducing CEVA-XM4™
4th-generation imaging and vision processor IP
Bringing embedded systems closer to human vision and visual
perception
Vector-type processor; combines fixed- and floating-point math; up to
4096-bit processing per cycle
Platform includes vision processor, libraries, tools and applications
Leverages infrastructure, tools and ecosystem of CEVA-MM3101, with
over a dozen licensees and 30 partners
Enables sophisticated computer vision algorithms including:
1. Real-time 3D depth mapping and point cloud processing
2. Computational photography for image enhancement algorithms
3. Deep learning and convolutional neural network (CNN) for object
detection, recognition and context-aware algorithms
Performance gain*
Up
to 8x
35%
Power savings*
* vs. CEVA-MM3101
3CEVA Proprietary Information
Some Vision Terminology
3D vision
Computational
photography
Visual
perception
CEVA-XM4 - Intelligent Vision Processing
- Image signal processor
(ISP)*
- Image registration
- Temporal noise reduction
- Depth map generation
- Refocus image
- Video stabilization
- Low-light image enhance
- Zoom
- Super-resolution
- Background removal
- HDR
- Object recognition
- Object tracking
- Augmented reality (AR)
- Natural user interface (NUI)
Left Image
Right Image
Depth Data
Images, Data
Encode*
* these are most appropriately implemented by HW accelerators
4CEVA Proprietary Information
CEVA-XM4 Block Diagram
Program Control
Unit
CEVA-Connect
AXI Masters and Slaves
AXI Master and Slave
Sequencer
Interrupts
Load Store
Unit
Dispatcher
Program MSS
Cache
Controller
Program DMA
AXI Master
Emulation
On-chip
Emulation
(OCEM)
Profiler
Real-time
Trace
Data MSS
Message
Queues
Data DMA
AXI Master
L1 Program
Memory
DMA Queue
Manager
AXI Master
TCE
User-defined
coprocessors
L1 Data
Memory
PSU
Vector Processing Unit
Vector Register File
VPU1VPU0
Vector
FPU
ISA
XTEND
LS0
LS1
Scalar
Processing Unit
Global Register File
SPU0 SPU1
SPU2 SPU3
Parallel
Random Mem
Access
Multi-Core
Interfaces
User-defined
HW accelerators
Automatic
management of
HW accelerators
128 MACs
Vector
Floating-Point
Specialized
DMA
4-Way Instruction
Cache
Scalar
Floating-Point
5CEVA Proprietary Information
• Face Detection & Recognition
• Emotion Recognition
• Gesture Recognition
• ADAS Algorithms (FCW, LDW)
• 3D Depth Map Creation
CEVA-XM4 Platform Layers
CPU-DSP Link – Communication Layer
• Digital Video Stabilizer (DVS)
• Super-Resolution (SR)
Hardware
Layer
Software
Layer
App Dev.
Kit (ADK)
Host CEVA-CV API / OpenVX
SW
Toolset
Hardware
Development
Kit
Partner Software Products
CEVA-XM4 Imaging & Vision Processor
SmartFrame – Automatic handling of system & memory transfers
CEVA Software Products
RTOS
CEVA-CV LibrariesAndroid Framework (AMF) Provides OEM
differentiation
CPU
offload
Source code
provided
6CEVA Proprietary Information
CEVA-XM4 Target Markets
Target devices and applications: smartphones,
tablets, automotive safety and infotainment, robotics,
security and surveillance, augmented reality, drones,
ADAS, autonomous driving, HMI/NUI, AEB
(Autonomous Emergency Breaking), Virtual Reality,
IoT, image recognition, HDR
Computer vision algorithms supported include real-
time 3D depth map generation and point cloud
processing for 3D scanning, object detection and
recognition algorithms, ranging from ORB, Haar, and
LBP, all the way to deep learning algorithms that use
neural network technologies such as convolutional
neural networks (CNN).
Computational Photography algorithms supported
include refocus, background replacement, zoom,
super-resolution, image stabilization, noise reduction
and improved low-light capabilities.
7CEVA Proprietary Information
CEVA-XM4 - a World of Possibilities
With CEVA-XM4, you can:
Run computer vision on video streams
(1080p, 4K)
Combine depth generation with vision
processing (e.g. depth + AR, depth +
3D scanning)
Enable multi-app processing (e.g.
gesture + face detection + emotions +
eye-tracking + optional depth)
Implement multi-image algorithms in
high resolution (e.g. super-resolution,
refocus)
Refocus Image
Low-light Image
Enhancement
Video Stabilization
3D Scanning
Depth Mapping
3D Registration
Augmented
Reality
Point Cloud
Emotion and Intent
Detection
Segmentation and
Classification
Deep Learning
(CNN)
Object Tracking
Object Recognition
Super Resolution
Digital Zoom
Computational
Photography
8CEVA Proprietary Information
CEVA-ToolBox™ - Eclipse based SW Dev. Tools
Advanced Eclipse-based IDE
Optimizing C/C++ compiler
Auto Vectorization
Extensive Vec-C support
C language extensions in OpenCL-like syntax
Vector types for C programming - short8,
ushort32,…
Vectorization from C operators
Linker and Utilities
Automatic Build Optimizer
Built-in Debugger & Simulator
C-level functional Profiler
Target Emulation dev. kit
Complete Software Development tools.
Focused on ease of use and quick SW porting for performance optimization
IDE / Debugger
View
Build
Optimizer
Profiler –
Function Graph
9CEVA Proprietary Information
CEVA Application Developer Kit (ADK)
► Via CEVA-CV extensive OpenCV-based library
► API directly to CPU development domain
► Performance- and power-optimized implementation
► Via SmartFrame tool auto handling of data transfer,
frame, tile
► Kernel execution with tunneling
► CPU-DSP communication channels
► Supports Android direct connection (AMF)
► Leverage existing pre-optimized libraries & algorithms
Abstracts CPU-DSP interface Abstracts DSP ISA
Accelerates CV algorithm development Abstracts all system and memory aspects
A broad set of tools designed to simplify overall software development and save
significantly on power consumption and memory bandwidth
Proven and mature
development
environment
10CEVA Proprietary Information
Summary
The era of “intelligent vision” is upon
us
Applicable to every conceivable end-
market and device that can be camera
enabled
CEVA-XM4 makes intelligent vision in
embedded systems a reality
Overcomes the processing, power and
die-size limitations inherent to
embedded systems
Now, the possibilities for ubiquitous
intelligent vision within our every day
devices is endless!
*http://www.fastcompany.com/1623012/will-smart-contact-lenses-be-bluetooth-headsets-future
Smart Contact Lens
For more information, visit www.ceva-dsp.com/XM4

Contenu connexe

En vedette

Back-propagation Primer
Back-propagation PrimerBack-propagation Primer
Back-propagation PrimerAuro Tripathy
 
CapGemini, Panalpina & Oracle on 'The Hidden Opportunities of Economic Crisis...
CapGemini, Panalpina & Oracle on 'The Hidden Opportunities of Economic Crisis...CapGemini, Panalpina & Oracle on 'The Hidden Opportunities of Economic Crisis...
CapGemini, Panalpina & Oracle on 'The Hidden Opportunities of Economic Crisis...eyefortransport
 
Machine Learning with New Hardware Challegens
Machine Learning with New Hardware ChallegensMachine Learning with New Hardware Challegens
Machine Learning with New Hardware ChallegensOscar Law
 
"Trade-offs in Implementing Deep Neural Networks on FPGAs," a Presentation fr...
"Trade-offs in Implementing Deep Neural Networks on FPGAs," a Presentation fr..."Trade-offs in Implementing Deep Neural Networks on FPGAs," a Presentation fr...
"Trade-offs in Implementing Deep Neural Networks on FPGAs," a Presentation fr...Edge AI and Vision Alliance
 
Face recognition using neural network
Face recognition using neural networkFace recognition using neural network
Face recognition using neural networkIndira Nayak
 
[251] implementing deep learning using cu dnn
[251] implementing deep learning using cu dnn[251] implementing deep learning using cu dnn
[251] implementing deep learning using cu dnnNAVER D2
 
Face recognition using artificial neural network
Face recognition using artificial neural networkFace recognition using artificial neural network
Face recognition using artificial neural networkSumeet Kakani
 
Artificial intelligence NEURAL NETWORKS
Artificial intelligence NEURAL NETWORKSArtificial intelligence NEURAL NETWORKS
Artificial intelligence NEURAL NETWORKSREHMAT ULLAH
 
neural network
neural networkneural network
neural networkSTUDENT
 
A Three-Dimensional Representation method for Noisy Point Clouds based on Gro...
A Three-Dimensional Representation method for Noisy Point Clouds based on Gro...A Three-Dimensional Representation method for Noisy Point Clouds based on Gro...
A Three-Dimensional Representation method for Noisy Point Clouds based on Gro...Sergio Orts-Escolano
 
Artificial neural network
Artificial neural networkArtificial neural network
Artificial neural networkmustafa aadel
 
Neural network & its applications
Neural network & its applications Neural network & its applications
Neural network & its applications Ahmed_hashmi
 
最近のDeep Learning (NLP) 界隈におけるAttention事情
最近のDeep Learning (NLP) 界隈におけるAttention事情最近のDeep Learning (NLP) 界隈におけるAttention事情
最近のDeep Learning (NLP) 界隈におけるAttention事情Yuta Kikuchi
 
Backpropagation in Convolutional Neural Network
Backpropagation in Convolutional Neural NetworkBackpropagation in Convolutional Neural Network
Backpropagation in Convolutional Neural NetworkHiroshi Kuwajima
 
Artificial neural networks
Artificial neural networksArtificial neural networks
Artificial neural networksstellajoseph
 
Artificial neural network
Artificial neural networkArtificial neural network
Artificial neural networkDEEPASHRI HK
 

En vedette (16)

Back-propagation Primer
Back-propagation PrimerBack-propagation Primer
Back-propagation Primer
 
CapGemini, Panalpina & Oracle on 'The Hidden Opportunities of Economic Crisis...
CapGemini, Panalpina & Oracle on 'The Hidden Opportunities of Economic Crisis...CapGemini, Panalpina & Oracle on 'The Hidden Opportunities of Economic Crisis...
CapGemini, Panalpina & Oracle on 'The Hidden Opportunities of Economic Crisis...
 
Machine Learning with New Hardware Challegens
Machine Learning with New Hardware ChallegensMachine Learning with New Hardware Challegens
Machine Learning with New Hardware Challegens
 
"Trade-offs in Implementing Deep Neural Networks on FPGAs," a Presentation fr...
"Trade-offs in Implementing Deep Neural Networks on FPGAs," a Presentation fr..."Trade-offs in Implementing Deep Neural Networks on FPGAs," a Presentation fr...
"Trade-offs in Implementing Deep Neural Networks on FPGAs," a Presentation fr...
 
Face recognition using neural network
Face recognition using neural networkFace recognition using neural network
Face recognition using neural network
 
[251] implementing deep learning using cu dnn
[251] implementing deep learning using cu dnn[251] implementing deep learning using cu dnn
[251] implementing deep learning using cu dnn
 
Face recognition using artificial neural network
Face recognition using artificial neural networkFace recognition using artificial neural network
Face recognition using artificial neural network
 
Artificial intelligence NEURAL NETWORKS
Artificial intelligence NEURAL NETWORKSArtificial intelligence NEURAL NETWORKS
Artificial intelligence NEURAL NETWORKS
 
neural network
neural networkneural network
neural network
 
A Three-Dimensional Representation method for Noisy Point Clouds based on Gro...
A Three-Dimensional Representation method for Noisy Point Clouds based on Gro...A Three-Dimensional Representation method for Noisy Point Clouds based on Gro...
A Three-Dimensional Representation method for Noisy Point Clouds based on Gro...
 
Artificial neural network
Artificial neural networkArtificial neural network
Artificial neural network
 
Neural network & its applications
Neural network & its applications Neural network & its applications
Neural network & its applications
 
最近のDeep Learning (NLP) 界隈におけるAttention事情
最近のDeep Learning (NLP) 界隈におけるAttention事情最近のDeep Learning (NLP) 界隈におけるAttention事情
最近のDeep Learning (NLP) 界隈におけるAttention事情
 
Backpropagation in Convolutional Neural Network
Backpropagation in Convolutional Neural NetworkBackpropagation in Convolutional Neural Network
Backpropagation in Convolutional Neural Network
 
Artificial neural networks
Artificial neural networksArtificial neural networks
Artificial neural networks
 
Artificial neural network
Artificial neural networkArtificial neural network
Artificial neural network
 

Plus de CEVA, Inc.

CEVA Q4 and Year End 2018 EarningsInfographic
CEVA Q4 and Year End 2018 EarningsInfographicCEVA Q4 and Year End 2018 EarningsInfographic
CEVA Q4 and Year End 2018 EarningsInfographicCEVA, Inc.
 
CEVA Q3 2018 Earnings Infographic
CEVA Q3 2018 Earnings InfographicCEVA Q3 2018 Earnings Infographic
CEVA Q3 2018 Earnings InfographicCEVA, Inc.
 
CEVA Q1 2018 Earnings Infographic
CEVA Q1 2018 Earnings InfographicCEVA Q1 2018 Earnings Infographic
CEVA Q1 2018 Earnings InfographicCEVA, Inc.
 
CEVA PentaG - 5G New Radio IP Platform for Enhanced Mobile Broadband
CEVA PentaG - 5G New Radio IP Platform for Enhanced Mobile BroadbandCEVA PentaG - 5G New Radio IP Platform for Enhanced Mobile Broadband
CEVA PentaG - 5G New Radio IP Platform for Enhanced Mobile BroadbandCEVA, Inc.
 
CEVA Q4 and Year End 2017 Earnings Infographic
CEVA Q4 and Year End 2017 Earnings InfographicCEVA Q4 and Year End 2017 Earnings Infographic
CEVA Q4 and Year End 2017 Earnings InfographicCEVA, Inc.
 
CEVA Q3 2017 Earnings Infographic
CEVA Q3 2017 Earnings InfographicCEVA Q3 2017 Earnings Infographic
CEVA Q3 2017 Earnings InfographicCEVA, Inc.
 
CEVA Q1 2017 Earnings Infographic
CEVA Q1 2017 Earnings InfographicCEVA Q1 2017 Earnings Infographic
CEVA Q1 2017 Earnings InfographicCEVA, Inc.
 
CEVA, Inc. Q2 2016 Earnings Infographic
CEVA, Inc. Q2 2016 Earnings InfographicCEVA, Inc. Q2 2016 Earnings Infographic
CEVA, Inc. Q2 2016 Earnings InfographicCEVA, Inc.
 
CEVA, Inc. Q1 2016 Earnings Infographic
CEVA, Inc. Q1 2016 Earnings InfographicCEVA, Inc. Q1 2016 Earnings Infographic
CEVA, Inc. Q1 2016 Earnings InfographicCEVA, Inc.
 
Q4 and Year End 2015 Earnings Infographic
Q4 and Year End 2015 Earnings InfographicQ4 and Year End 2015 Earnings Infographic
Q4 and Year End 2015 Earnings InfographicCEVA, Inc.
 
CEVA, Inc. Q3 2015 Earnings Infographic
CEVA, Inc. Q3 2015 Earnings InfographicCEVA, Inc. Q3 2015 Earnings Infographic
CEVA, Inc. Q3 2015 Earnings InfographicCEVA, Inc.
 
CEVA, Inc. Q2 2015 Earnings Infographic
CEVA, Inc. Q2 2015 Earnings InfographicCEVA, Inc. Q2 2015 Earnings Infographic
CEVA, Inc. Q2 2015 Earnings InfographicCEVA, Inc.
 
CEVA, Inc. Q1 2015 Earnings Infographic
CEVA, Inc. Q1 2015 Earnings InfographicCEVA, Inc. Q1 2015 Earnings Infographic
CEVA, Inc. Q1 2015 Earnings InfographicCEVA, Inc.
 

Plus de CEVA, Inc. (13)

CEVA Q4 and Year End 2018 EarningsInfographic
CEVA Q4 and Year End 2018 EarningsInfographicCEVA Q4 and Year End 2018 EarningsInfographic
CEVA Q4 and Year End 2018 EarningsInfographic
 
CEVA Q3 2018 Earnings Infographic
CEVA Q3 2018 Earnings InfographicCEVA Q3 2018 Earnings Infographic
CEVA Q3 2018 Earnings Infographic
 
CEVA Q1 2018 Earnings Infographic
CEVA Q1 2018 Earnings InfographicCEVA Q1 2018 Earnings Infographic
CEVA Q1 2018 Earnings Infographic
 
CEVA PentaG - 5G New Radio IP Platform for Enhanced Mobile Broadband
CEVA PentaG - 5G New Radio IP Platform for Enhanced Mobile BroadbandCEVA PentaG - 5G New Radio IP Platform for Enhanced Mobile Broadband
CEVA PentaG - 5G New Radio IP Platform for Enhanced Mobile Broadband
 
CEVA Q4 and Year End 2017 Earnings Infographic
CEVA Q4 and Year End 2017 Earnings InfographicCEVA Q4 and Year End 2017 Earnings Infographic
CEVA Q4 and Year End 2017 Earnings Infographic
 
CEVA Q3 2017 Earnings Infographic
CEVA Q3 2017 Earnings InfographicCEVA Q3 2017 Earnings Infographic
CEVA Q3 2017 Earnings Infographic
 
CEVA Q1 2017 Earnings Infographic
CEVA Q1 2017 Earnings InfographicCEVA Q1 2017 Earnings Infographic
CEVA Q1 2017 Earnings Infographic
 
CEVA, Inc. Q2 2016 Earnings Infographic
CEVA, Inc. Q2 2016 Earnings InfographicCEVA, Inc. Q2 2016 Earnings Infographic
CEVA, Inc. Q2 2016 Earnings Infographic
 
CEVA, Inc. Q1 2016 Earnings Infographic
CEVA, Inc. Q1 2016 Earnings InfographicCEVA, Inc. Q1 2016 Earnings Infographic
CEVA, Inc. Q1 2016 Earnings Infographic
 
Q4 and Year End 2015 Earnings Infographic
Q4 and Year End 2015 Earnings InfographicQ4 and Year End 2015 Earnings Infographic
Q4 and Year End 2015 Earnings Infographic
 
CEVA, Inc. Q3 2015 Earnings Infographic
CEVA, Inc. Q3 2015 Earnings InfographicCEVA, Inc. Q3 2015 Earnings Infographic
CEVA, Inc. Q3 2015 Earnings Infographic
 
CEVA, Inc. Q2 2015 Earnings Infographic
CEVA, Inc. Q2 2015 Earnings InfographicCEVA, Inc. Q2 2015 Earnings Infographic
CEVA, Inc. Q2 2015 Earnings Infographic
 
CEVA, Inc. Q1 2015 Earnings Infographic
CEVA, Inc. Q1 2015 Earnings InfographicCEVA, Inc. Q1 2015 Earnings Infographic
CEVA, Inc. Q1 2015 Earnings Infographic
 

Dernier

COMPUTER 10: Lesson 7 - File Storage and Online Collaboration
COMPUTER 10: Lesson 7 - File Storage and Online CollaborationCOMPUTER 10: Lesson 7 - File Storage and Online Collaboration
COMPUTER 10: Lesson 7 - File Storage and Online Collaborationbruanjhuli
 
Meet the new FSP 3000 M-Flex800™
Meet the new FSP 3000 M-Flex800™Meet the new FSP 3000 M-Flex800™
Meet the new FSP 3000 M-Flex800™Adtran
 
Computer 10: Lesson 10 - Online Crimes and Hazards
Computer 10: Lesson 10 - Online Crimes and HazardsComputer 10: Lesson 10 - Online Crimes and Hazards
Computer 10: Lesson 10 - Online Crimes and HazardsSeth Reyes
 
Artificial Intelligence & SEO Trends for 2024
Artificial Intelligence & SEO Trends for 2024Artificial Intelligence & SEO Trends for 2024
Artificial Intelligence & SEO Trends for 2024D Cloud Solutions
 
Cloud Revolution: Exploring the New Wave of Serverless Spatial Data
Cloud Revolution: Exploring the New Wave of Serverless Spatial DataCloud Revolution: Exploring the New Wave of Serverless Spatial Data
Cloud Revolution: Exploring the New Wave of Serverless Spatial DataSafe Software
 
UiPath Solutions Management Preview - Northern CA Chapter - March 22.pdf
UiPath Solutions Management Preview - Northern CA Chapter - March 22.pdfUiPath Solutions Management Preview - Northern CA Chapter - March 22.pdf
UiPath Solutions Management Preview - Northern CA Chapter - March 22.pdfDianaGray10
 
Do we need a new standard for visualizing the invisible?
Do we need a new standard for visualizing the invisible?Do we need a new standard for visualizing the invisible?
Do we need a new standard for visualizing the invisible?SANGHEE SHIN
 
The Data Metaverse: Unpacking the Roles, Use Cases, and Tech Trends in Data a...
The Data Metaverse: Unpacking the Roles, Use Cases, and Tech Trends in Data a...The Data Metaverse: Unpacking the Roles, Use Cases, and Tech Trends in Data a...
The Data Metaverse: Unpacking the Roles, Use Cases, and Tech Trends in Data a...Aggregage
 
Videogame localization & technology_ how to enhance the power of translation.pdf
Videogame localization & technology_ how to enhance the power of translation.pdfVideogame localization & technology_ how to enhance the power of translation.pdf
Videogame localization & technology_ how to enhance the power of translation.pdfinfogdgmi
 
Apres-Cyber - The Data Dilemma: Bridging Offensive Operations and Machine Lea...
Apres-Cyber - The Data Dilemma: Bridging Offensive Operations and Machine Lea...Apres-Cyber - The Data Dilemma: Bridging Offensive Operations and Machine Lea...
Apres-Cyber - The Data Dilemma: Bridging Offensive Operations and Machine Lea...Will Schroeder
 
Empowering Africa's Next Generation: The AI Leadership Blueprint
Empowering Africa's Next Generation: The AI Leadership BlueprintEmpowering Africa's Next Generation: The AI Leadership Blueprint
Empowering Africa's Next Generation: The AI Leadership BlueprintMahmoud Rabie
 
UiPath Community: AI for UiPath Automation Developers
UiPath Community: AI for UiPath Automation DevelopersUiPath Community: AI for UiPath Automation Developers
UiPath Community: AI for UiPath Automation DevelopersUiPathCommunity
 
Basic Building Blocks of Internet of Things.
Basic Building Blocks of Internet of Things.Basic Building Blocks of Internet of Things.
Basic Building Blocks of Internet of Things.YounusS2
 
Anypoint Code Builder , Google Pub sub connector and MuleSoft RPA
Anypoint Code Builder , Google Pub sub connector and MuleSoft RPAAnypoint Code Builder , Google Pub sub connector and MuleSoft RPA
Anypoint Code Builder , Google Pub sub connector and MuleSoft RPAshyamraj55
 
9 Steps For Building Winning Founding Team
9 Steps For Building Winning Founding Team9 Steps For Building Winning Founding Team
9 Steps For Building Winning Founding TeamAdam Moalla
 
Machine Learning Model Validation (Aijun Zhang 2024).pdf
Machine Learning Model Validation (Aijun Zhang 2024).pdfMachine Learning Model Validation (Aijun Zhang 2024).pdf
Machine Learning Model Validation (Aijun Zhang 2024).pdfAijun Zhang
 
Introduction to Matsuo Laboratory (ENG).pptx
Introduction to Matsuo Laboratory (ENG).pptxIntroduction to Matsuo Laboratory (ENG).pptx
Introduction to Matsuo Laboratory (ENG).pptxMatsuo Lab
 
Nanopower In Semiconductor Industry.pdf
Nanopower  In Semiconductor Industry.pdfNanopower  In Semiconductor Industry.pdf
Nanopower In Semiconductor Industry.pdfPedro Manuel
 
Linked Data in Production: Moving Beyond Ontologies
Linked Data in Production: Moving Beyond OntologiesLinked Data in Production: Moving Beyond Ontologies
Linked Data in Production: Moving Beyond OntologiesDavid Newbury
 
Things you didn't know you can use in your Salesforce
Things you didn't know you can use in your SalesforceThings you didn't know you can use in your Salesforce
Things you didn't know you can use in your SalesforceMartin Humpolec
 

Dernier (20)

COMPUTER 10: Lesson 7 - File Storage and Online Collaboration
COMPUTER 10: Lesson 7 - File Storage and Online CollaborationCOMPUTER 10: Lesson 7 - File Storage and Online Collaboration
COMPUTER 10: Lesson 7 - File Storage and Online Collaboration
 
Meet the new FSP 3000 M-Flex800™
Meet the new FSP 3000 M-Flex800™Meet the new FSP 3000 M-Flex800™
Meet the new FSP 3000 M-Flex800™
 
Computer 10: Lesson 10 - Online Crimes and Hazards
Computer 10: Lesson 10 - Online Crimes and HazardsComputer 10: Lesson 10 - Online Crimes and Hazards
Computer 10: Lesson 10 - Online Crimes and Hazards
 
Artificial Intelligence & SEO Trends for 2024
Artificial Intelligence & SEO Trends for 2024Artificial Intelligence & SEO Trends for 2024
Artificial Intelligence & SEO Trends for 2024
 
Cloud Revolution: Exploring the New Wave of Serverless Spatial Data
Cloud Revolution: Exploring the New Wave of Serverless Spatial DataCloud Revolution: Exploring the New Wave of Serverless Spatial Data
Cloud Revolution: Exploring the New Wave of Serverless Spatial Data
 
UiPath Solutions Management Preview - Northern CA Chapter - March 22.pdf
UiPath Solutions Management Preview - Northern CA Chapter - March 22.pdfUiPath Solutions Management Preview - Northern CA Chapter - March 22.pdf
UiPath Solutions Management Preview - Northern CA Chapter - March 22.pdf
 
Do we need a new standard for visualizing the invisible?
Do we need a new standard for visualizing the invisible?Do we need a new standard for visualizing the invisible?
Do we need a new standard for visualizing the invisible?
 
The Data Metaverse: Unpacking the Roles, Use Cases, and Tech Trends in Data a...
The Data Metaverse: Unpacking the Roles, Use Cases, and Tech Trends in Data a...The Data Metaverse: Unpacking the Roles, Use Cases, and Tech Trends in Data a...
The Data Metaverse: Unpacking the Roles, Use Cases, and Tech Trends in Data a...
 
Videogame localization & technology_ how to enhance the power of translation.pdf
Videogame localization & technology_ how to enhance the power of translation.pdfVideogame localization & technology_ how to enhance the power of translation.pdf
Videogame localization & technology_ how to enhance the power of translation.pdf
 
Apres-Cyber - The Data Dilemma: Bridging Offensive Operations and Machine Lea...
Apres-Cyber - The Data Dilemma: Bridging Offensive Operations and Machine Lea...Apres-Cyber - The Data Dilemma: Bridging Offensive Operations and Machine Lea...
Apres-Cyber - The Data Dilemma: Bridging Offensive Operations and Machine Lea...
 
Empowering Africa's Next Generation: The AI Leadership Blueprint
Empowering Africa's Next Generation: The AI Leadership BlueprintEmpowering Africa's Next Generation: The AI Leadership Blueprint
Empowering Africa's Next Generation: The AI Leadership Blueprint
 
UiPath Community: AI for UiPath Automation Developers
UiPath Community: AI for UiPath Automation DevelopersUiPath Community: AI for UiPath Automation Developers
UiPath Community: AI for UiPath Automation Developers
 
Basic Building Blocks of Internet of Things.
Basic Building Blocks of Internet of Things.Basic Building Blocks of Internet of Things.
Basic Building Blocks of Internet of Things.
 
Anypoint Code Builder , Google Pub sub connector and MuleSoft RPA
Anypoint Code Builder , Google Pub sub connector and MuleSoft RPAAnypoint Code Builder , Google Pub sub connector and MuleSoft RPA
Anypoint Code Builder , Google Pub sub connector and MuleSoft RPA
 
9 Steps For Building Winning Founding Team
9 Steps For Building Winning Founding Team9 Steps For Building Winning Founding Team
9 Steps For Building Winning Founding Team
 
Machine Learning Model Validation (Aijun Zhang 2024).pdf
Machine Learning Model Validation (Aijun Zhang 2024).pdfMachine Learning Model Validation (Aijun Zhang 2024).pdf
Machine Learning Model Validation (Aijun Zhang 2024).pdf
 
Introduction to Matsuo Laboratory (ENG).pptx
Introduction to Matsuo Laboratory (ENG).pptxIntroduction to Matsuo Laboratory (ENG).pptx
Introduction to Matsuo Laboratory (ENG).pptx
 
Nanopower In Semiconductor Industry.pdf
Nanopower  In Semiconductor Industry.pdfNanopower  In Semiconductor Industry.pdf
Nanopower In Semiconductor Industry.pdf
 
Linked Data in Production: Moving Beyond Ontologies
Linked Data in Production: Moving Beyond OntologiesLinked Data in Production: Moving Beyond Ontologies
Linked Data in Production: Moving Beyond Ontologies
 
Things you didn't know you can use in your Salesforce
Things you didn't know you can use in your SalesforceThings you didn't know you can use in your Salesforce
Things you didn't know you can use in your Salesforce
 

CEVA-XM4 - The Intelligent Vision Processor IP

  • 1. Bringing Intelligent Vision Processing to Low-Power Embedded Systems February 2015
  • 2. 2CEVA Proprietary Information Introducing CEVA-XM4™ 4th-generation imaging and vision processor IP Bringing embedded systems closer to human vision and visual perception Vector-type processor; combines fixed- and floating-point math; up to 4096-bit processing per cycle Platform includes vision processor, libraries, tools and applications Leverages infrastructure, tools and ecosystem of CEVA-MM3101, with over a dozen licensees and 30 partners Enables sophisticated computer vision algorithms including: 1. Real-time 3D depth mapping and point cloud processing 2. Computational photography for image enhancement algorithms 3. Deep learning and convolutional neural network (CNN) for object detection, recognition and context-aware algorithms Performance gain* Up to 8x 35% Power savings* * vs. CEVA-MM3101
  • 3. 3CEVA Proprietary Information Some Vision Terminology 3D vision Computational photography Visual perception CEVA-XM4 - Intelligent Vision Processing - Image signal processor (ISP)* - Image registration - Temporal noise reduction - Depth map generation - Refocus image - Video stabilization - Low-light image enhance - Zoom - Super-resolution - Background removal - HDR - Object recognition - Object tracking - Augmented reality (AR) - Natural user interface (NUI) Left Image Right Image Depth Data Images, Data Encode* * these are most appropriately implemented by HW accelerators
  • 4. 4CEVA Proprietary Information CEVA-XM4 Block Diagram Program Control Unit CEVA-Connect AXI Masters and Slaves AXI Master and Slave Sequencer Interrupts Load Store Unit Dispatcher Program MSS Cache Controller Program DMA AXI Master Emulation On-chip Emulation (OCEM) Profiler Real-time Trace Data MSS Message Queues Data DMA AXI Master L1 Program Memory DMA Queue Manager AXI Master TCE User-defined coprocessors L1 Data Memory PSU Vector Processing Unit Vector Register File VPU1VPU0 Vector FPU ISA XTEND LS0 LS1 Scalar Processing Unit Global Register File SPU0 SPU1 SPU2 SPU3 Parallel Random Mem Access Multi-Core Interfaces User-defined HW accelerators Automatic management of HW accelerators 128 MACs Vector Floating-Point Specialized DMA 4-Way Instruction Cache Scalar Floating-Point
  • 5. 5CEVA Proprietary Information • Face Detection & Recognition • Emotion Recognition • Gesture Recognition • ADAS Algorithms (FCW, LDW) • 3D Depth Map Creation CEVA-XM4 Platform Layers CPU-DSP Link – Communication Layer • Digital Video Stabilizer (DVS) • Super-Resolution (SR) Hardware Layer Software Layer App Dev. Kit (ADK) Host CEVA-CV API / OpenVX SW Toolset Hardware Development Kit Partner Software Products CEVA-XM4 Imaging & Vision Processor SmartFrame – Automatic handling of system & memory transfers CEVA Software Products RTOS CEVA-CV LibrariesAndroid Framework (AMF) Provides OEM differentiation CPU offload Source code provided
  • 6. 6CEVA Proprietary Information CEVA-XM4 Target Markets Target devices and applications: smartphones, tablets, automotive safety and infotainment, robotics, security and surveillance, augmented reality, drones, ADAS, autonomous driving, HMI/NUI, AEB (Autonomous Emergency Breaking), Virtual Reality, IoT, image recognition, HDR Computer vision algorithms supported include real- time 3D depth map generation and point cloud processing for 3D scanning, object detection and recognition algorithms, ranging from ORB, Haar, and LBP, all the way to deep learning algorithms that use neural network technologies such as convolutional neural networks (CNN). Computational Photography algorithms supported include refocus, background replacement, zoom, super-resolution, image stabilization, noise reduction and improved low-light capabilities.
  • 7. 7CEVA Proprietary Information CEVA-XM4 - a World of Possibilities With CEVA-XM4, you can: Run computer vision on video streams (1080p, 4K) Combine depth generation with vision processing (e.g. depth + AR, depth + 3D scanning) Enable multi-app processing (e.g. gesture + face detection + emotions + eye-tracking + optional depth) Implement multi-image algorithms in high resolution (e.g. super-resolution, refocus) Refocus Image Low-light Image Enhancement Video Stabilization 3D Scanning Depth Mapping 3D Registration Augmented Reality Point Cloud Emotion and Intent Detection Segmentation and Classification Deep Learning (CNN) Object Tracking Object Recognition Super Resolution Digital Zoom Computational Photography
  • 8. 8CEVA Proprietary Information CEVA-ToolBox™ - Eclipse based SW Dev. Tools Advanced Eclipse-based IDE Optimizing C/C++ compiler Auto Vectorization Extensive Vec-C support C language extensions in OpenCL-like syntax Vector types for C programming - short8, ushort32,… Vectorization from C operators Linker and Utilities Automatic Build Optimizer Built-in Debugger & Simulator C-level functional Profiler Target Emulation dev. kit Complete Software Development tools. Focused on ease of use and quick SW porting for performance optimization IDE / Debugger View Build Optimizer Profiler – Function Graph
  • 9. 9CEVA Proprietary Information CEVA Application Developer Kit (ADK) ► Via CEVA-CV extensive OpenCV-based library ► API directly to CPU development domain ► Performance- and power-optimized implementation ► Via SmartFrame tool auto handling of data transfer, frame, tile ► Kernel execution with tunneling ► CPU-DSP communication channels ► Supports Android direct connection (AMF) ► Leverage existing pre-optimized libraries & algorithms Abstracts CPU-DSP interface Abstracts DSP ISA Accelerates CV algorithm development Abstracts all system and memory aspects A broad set of tools designed to simplify overall software development and save significantly on power consumption and memory bandwidth Proven and mature development environment
  • 10. 10CEVA Proprietary Information Summary The era of “intelligent vision” is upon us Applicable to every conceivable end- market and device that can be camera enabled CEVA-XM4 makes intelligent vision in embedded systems a reality Overcomes the processing, power and die-size limitations inherent to embedded systems Now, the possibilities for ubiquitous intelligent vision within our every day devices is endless! *http://www.fastcompany.com/1623012/will-smart-contact-lenses-be-bluetooth-headsets-future Smart Contact Lens
  • 11. For more information, visit www.ceva-dsp.com/XM4