SlideShare a Scribd company logo
1 of 46
Design considerations for IoT systems
Yossi Cohen
December 2020
Outline
• IoT Modern needs
• Deep Learning networks and operations
• CPU GPU TPU and NN
• Considerations
• Dev-boards and NPU accelerators
– RPI
– NVIDIA
– Rockchip
– Intel
• Cameras 2D / 3D
• Summary
IOT NEEDS
Basic Edge Processor operation
• Worked as controller
• Control motors / open gates
• Take pictures / basic image operations
• Send images to the cloud for further
processing/storage/decision
• Operate based on cloud decision
IoT Modern roles
• Operations done in IoT device
– Read numbers, license plates, road signs
– Detect faces (dogs, cats)
• Count people
• Segment cars, pedestrians
• Scan environment / Navigate
• Drive cars / Fly drones
Neural Networks
• Basic idea: Multiply each input by weights
and sum
• Exists for 60+ years !!
– Implemented with resistors as weights
• Deep Neural networks 100+ layers
– Hundreds of millions or more weights
– Heavy matrix operations
LSTM/RNN
• Used for
– sound detectors (gun shot detector)
– data series - number of clients in restaurant
– Activity detector – based on accelerometers
– Usually requires low processing compared to
CNNs
CNN - Classifiers
• Used for
– Is object X in the image
– Is there a cat in the Image ?
– Common implementation LaNet
BOX Detectors
• Box detectors detects where is the object
located
• Used to detect:
– People, face, corona mask
– car, road signs
– cats
• Known box detectors:
– YOLO3 YOLO5
– SSD
Segmenter and Multi-Instance Segmenter
• Segmentor
– defines class for each pixel
– Pixel(X,Y) belongs to cat
– FCN-8 basic implementation
– Does not separate cats
• Multi-instance segmentor
– Separate Cats
– Pixel(X,Y) belongs to cat42
– MRCNN – common implementation
CPU and CNNs
• X86, ARM – sequential instruction
processing
• Not suitable for matrix (DL) operations !
• Remedies
• +SIMD – MMX SSE
• +Multicore 2,4,8….
• Yet….
Moore law is dead??
GPU – more than a graphic processor
• GPU are using 100s-1000s of (simple)
cores
• Better in parallel processing than CPU
• Much better for large matrix operations
13
GPU Examples
• GPU Cards
– NVIDIA 20XX
– NVIDIA T4
• Can we run a GPU only IoT solution?
TPUs
• TPUs are built ONLT for deep learning
• Designed for single operation layer
processing
• First GPUs started 4 years ago
• Google Coral, Hauwei, NVIDIA, Intel
Movidius, Narvana,
• In Israel: Habana, Heilo
• In most modern phones: Samsung apple,
QCOM
TPU
All internal buses are 256B wide
Processor selection is obvious right?
• TPU is the best solution for Deep learning
activities
DEV BOARDS AND
ACCELERATORS
Edge TPU accelerators
• Intel Movidius Processor
– Edge AI processor / accelerator
– Either chip or USB
– Connects to RPI
• Google TPU (Coral)
– Chip or USB
• Rockchip RK 1808
Processors today are SoC
• SoC which usually include:
– 4-8 ARM cores
– GPU (ARM Mali, NVIDIA XX, Intel XX)
– Sometimes a TPU
• The selection is CPU/GPU/TPU
• Selection of solution is between SoC
• For MVP or development phases selection is
between SoC development boards !!
Considerations
• Price - $$
• Performance TFlops/Sec
• Power - Watts
• Longevity
• Existing code base
• Development boards
• Connectivity
• SDKs
Price
• 10-20$ Arduino
• 35-75$ RPI3 / RPI4 + extended memory
• 59-100$ Jetson nano 2-4GB
• 100$ Intel Movidius accelerator
• 115$ Google Coral dev board
• 175$ Rockchip 3399 Dev board
• 200$ Jetson TX2
• 400$ Jetson NX
• 700$ Jetson XVG
Performance
• Performance ratio between Jetson
processors
Operating systems
• Usually, ubuntu running on most devices
• Android on some boards
• Raspbian / Raspberry Pi OS for RPI
Longevity
• Important to check the processor lifespan
• For example, in NVIDIA jetson processors
RPI3 / 4 dev boards
• 35-75$ depending on GB
• Most common dev board
• Control/GPIO/basic image processing
• ~3-6 fps of TinyYOLO
NVIDIA Jetson nano
• 59-100$ dev board
• Relatively strong GPU
• 25 fps TinyYOLO
Model Application Framework
NVIDIA Jetson
Nano Raspberry Pi 3
Tiny YOLO V3
(416×416)
Object Detection Darknet 25 FPS 0.5 FPS
OpenPose
(256×256)
Pose Estimation Caffe 14 FPS DNR
VGG-19
(224×224)
Classification MXNet 10 FPS 0.5 FPS
Super Resolution
(481×321)
Image Processing PyTorch 15 FPS DNR
Unet
(1x512x512)
Segmentation Caffe 18 FPS DNR
Source: https://developer.nvidia.com/embedded/jetson-nano-dl-inference-benchmarks
Rockchip 3399 dev board
• 175 – 299$ depending on board
• USB2/3
• PCIe connectivity
• GPIO
• TPU accelerated
Intel Atom
• UP – boards
• 200-700$ depending on kit including
RealSense for some options
• Use OpenVino SDK for development and
optimization
Jetson NX
• 399$ dev board
• X4-5 stronger than the nano
• Same RPI-like form factor
and connectivity
• emmc + PCIe
• Can do:
Software / open source
• RPI has many open-source solutions
• NVIDIA uses DeepStream and ISAAC SDKs
to accelerate development
• Intel uses OpenVino to accelerate
development on its boards
2D Camera
• MIPI / USB connectivity / HDMI
• Resolution / Formats YUV, JPEG, Bayer
• HDR / Pixel size
• Filters, IR Cut
• Lens quality, FOV, distortion
• ISP quality – very important for DL
• Not all cameras supported by all boards
• USB camera – UVC no need for driver on
Linux windows
Placing AI in the camera
• Many IoT solutions require depth camera
/ Lidar - Lots of processing
• To reduce processor cost done inside the
camera:
– Intel Realsense family
– Rockchip RMSL 2031000
– Veronica - Inutive
RealSense
• D455/D435/D415 cameras
– IR + RGB + Projector + dual cam
• Option for IMU for robots/drones
• On board processor
• Not fit for RPI in high-res(high data rate)
Inutive - Veronica
• Designed for robots  drones  AR
• Low latency to avoid AR motion sickness
• On chip SLAM
• Built in VPU
• Output depth map
ROCKchip RMSL203
• Rockchip depth camera solution
• Built in VPU/DSP
• Output depthmap
• Projector + IR + RGB
Placing AI in the camera
• Many IoT solutions require depth camera
/ Lidar - Lots of processing
• To reduce processor cost done inside the
camera:
– Intel Realsense family
– Rockchip RMSL 2031000
– Veronica - Inutive
RealSense
• D455/D435/D415 cameras
– IR + RGB + Projector + dual cam
• Option for IMU for robots/drones
• On board processor
• Not fit for RPI in high-res(high data rate)
Inutive - Veronica
• Designed for robots  drones  AR
• Low latency to avoid AR motion sickness
• On chip SLAM
• Built in VPU
• Output depth map
ROCKchip RMSL203
• Rockchip depth camera solution
• Built in VPU/DSP
• Output depthmap
• Projector + IR + RGB
Summary
• There isnt a wining solution
• Robotics / drones solutions differ from
video analytics
• First product MVP should prefer
development time & costs over
price/performance
• Select a product with good, well tried
SDKs and community
Cloud connectivity
• Why we need it?
– Heavy data/image/deep learning processing
– Data aggregation from many sensing points
– Logging
– Dashboard
– Health test
– Secure access
– OTA updates
• Both AWS and Azure has Cloud IoT solution which
seamless integrate with RPI3 and Jetson Nano
NVIDIA and AWS IoT (Core & Greengrass)
• Connecting to AWS IoT is easy
• Deepstream has already built in MQTT message
formatters and brokers designed for AWS
connectivity
Source: https://aws.amazon.com/blogs/iot/how-to-integrate-nvidia-deepstream-on-jetson-devices-with-aws-iot-core-and-aws-iot-
greengrass/
Cloud connectivity
• Why we need it?
– Heavy data/image/deep learning processing
– Data aggregation from many sensing points
– Logging
– Dashboard
– Health test
– Secure access
– OTA updates
• Both AWS and Azure has Cloud IoT solution which
seamless integrate with RPI3 Jetson products and
Intel products
Summary
• There isn't a wining solution
• Robotics / drones solutions differ from
video analytics
• First product MVP should prefer
development time & costs over
price/performance
• Select a product with good, well tried
SDKs and community
• Invest in a good camera / ISP
Contact
• Yossi Cohen CTO@DSP-IP
• +972-54-5313092 | yossi@dsp-ip.com
• What we do:
– Video Analytics
– Robotics
– Drones
– Autonomous cars
– Agritec

More Related Content

What's hot

MM-4085, Designing a game audio engine for HSA, by Laurent Betbeder
MM-4085, Designing a game audio engine for HSA, by Laurent BetbederMM-4085, Designing a game audio engine for HSA, by Laurent Betbeder
MM-4085, Designing a game audio engine for HSA, by Laurent BetbederAMD Developer Central
 
Jetson AGX Xavier and the New Era of Autonomous Machines
Jetson AGX Xavier and the New Era of Autonomous MachinesJetson AGX Xavier and the New Era of Autonomous Machines
Jetson AGX Xavier and the New Era of Autonomous MachinesDustin Franklin
 
Video Surveillance Using Raspberry Pi Architecture
Video Surveillance Using Raspberry Pi ArchitectureVideo Surveillance Using Raspberry Pi Architecture
Video Surveillance Using Raspberry Pi ArchitectureRUTURAJ SHETE
 
Data Science Week 2016. NVIDIA. "Платформы и инструменты для реализации систе...
Data Science Week 2016. NVIDIA. "Платформы и инструменты для реализации систе...Data Science Week 2016. NVIDIA. "Платформы и инструменты для реализации систе...
Data Science Week 2016. NVIDIA. "Платформы и инструменты для реализации систе...Newprolab
 
Qualcomm Vuforia Mobile Vision Platform: Smart Terrain for Depth-Sensing Cameras
Qualcomm Vuforia Mobile Vision Platform: Smart Terrain for Depth-Sensing CamerasQualcomm Vuforia Mobile Vision Platform: Smart Terrain for Depth-Sensing Cameras
Qualcomm Vuforia Mobile Vision Platform: Smart Terrain for Depth-Sensing CamerasQualcomm Developer Network
 
GS-4136, Optimizing Game Development using AMD’s GPU PerfStudio 2, by Gordon ...
GS-4136, Optimizing Game Development using AMD’s GPU PerfStudio 2, by Gordon ...GS-4136, Optimizing Game Development using AMD’s GPU PerfStudio 2, by Gordon ...
GS-4136, Optimizing Game Development using AMD’s GPU PerfStudio 2, by Gordon ...AMD Developer Central
 
Webinar: NVIDIA JETSON – A Inteligência Artificial na palma de sua mão
Webinar: NVIDIA JETSON – A Inteligência Artificial na palma de sua mãoWebinar: NVIDIA JETSON – A Inteligência Artificial na palma de sua mão
Webinar: NVIDIA JETSON – A Inteligência Artificial na palma de sua mãoEmbarcados
 
0xdroid -- community-developed Android distribution by 0xlab
0xdroid -- community-developed Android distribution by 0xlab0xdroid -- community-developed Android distribution by 0xlab
0xdroid -- community-developed Android distribution by 0xlabNational Cheng Kung University
 
Embedded and Reliable Computer Vision
Embedded and Reliable Computer VisionEmbedded and Reliable Computer Vision
Embedded and Reliable Computer VisionNVIDIA Taiwan
 
Accelerate graphics performance with ozone-gbm on Intel based Linux desktop s...
Accelerate graphics performance with ozone-gbm on Intel based Linux desktop s...Accelerate graphics performance with ozone-gbm on Intel based Linux desktop s...
Accelerate graphics performance with ozone-gbm on Intel based Linux desktop s...Joone Hur
 
Java fx on raspberry pi gerrit grunwald
Java fx on raspberry pi   gerrit grunwaldJava fx on raspberry pi   gerrit grunwald
Java fx on raspberry pi gerrit grunwaldNLJUG
 
Computer Vision Powered by Heterogeneous System Architecture (HSA) by Dr. Ha...
Computer Vision Powered by Heterogeneous System Architecture (HSA) by  Dr. Ha...Computer Vision Powered by Heterogeneous System Architecture (HSA) by  Dr. Ha...
Computer Vision Powered by Heterogeneous System Architecture (HSA) by Dr. Ha...AMD Developer Central
 
Qualcomm Hexagon SDK: Optimize Your Multimedia Solutions
Qualcomm Hexagon SDK: Optimize Your Multimedia SolutionsQualcomm Hexagon SDK: Optimize Your Multimedia Solutions
Qualcomm Hexagon SDK: Optimize Your Multimedia SolutionsQualcomm Developer Network
 
Digital Watchdog DWC-MD421DB Data Sheet
Digital Watchdog DWC-MD421DB Data SheetDigital Watchdog DWC-MD421DB Data Sheet
Digital Watchdog DWC-MD421DB Data SheetJMAC Supply
 
Choosing the Right IP Cores for Low-Latency Video Streaming
Choosing the Right IP Cores for Low-Latency Video StreamingChoosing the Right IP Cores for Low-Latency Video Streaming
Choosing the Right IP Cores for Low-Latency Video StreamingCAST, Inc.
 
Intel® Open Image Denoise: Optimized CPU Denoising | SIGGRAPH 2019 Technical ...
Intel® Open Image Denoise: Optimized CPU Denoising | SIGGRAPH 2019 Technical ...Intel® Open Image Denoise: Optimized CPU Denoising | SIGGRAPH 2019 Technical ...
Intel® Open Image Denoise: Optimized CPU Denoising | SIGGRAPH 2019 Technical ...Intel® Software
 

What's hot (20)

MM-4085, Designing a game audio engine for HSA, by Laurent Betbeder
MM-4085, Designing a game audio engine for HSA, by Laurent BetbederMM-4085, Designing a game audio engine for HSA, by Laurent Betbeder
MM-4085, Designing a game audio engine for HSA, by Laurent Betbeder
 
Radeon Vega Pro
Radeon Vega ProRadeon Vega Pro
Radeon Vega Pro
 
Jetson AGX Xavier and the New Era of Autonomous Machines
Jetson AGX Xavier and the New Era of Autonomous MachinesJetson AGX Xavier and the New Era of Autonomous Machines
Jetson AGX Xavier and the New Era of Autonomous Machines
 
Radeon RX Vega
Radeon RX Vega Radeon RX Vega
Radeon RX Vega
 
Video Surveillance Using Raspberry Pi Architecture
Video Surveillance Using Raspberry Pi ArchitectureVideo Surveillance Using Raspberry Pi Architecture
Video Surveillance Using Raspberry Pi Architecture
 
Data Science Week 2016. NVIDIA. "Платформы и инструменты для реализации систе...
Data Science Week 2016. NVIDIA. "Платформы и инструменты для реализации систе...Data Science Week 2016. NVIDIA. "Платформы и инструменты для реализации систе...
Data Science Week 2016. NVIDIA. "Платформы и инструменты для реализации систе...
 
Qualcomm Vuforia Mobile Vision Platform: Smart Terrain for Depth-Sensing Cameras
Qualcomm Vuforia Mobile Vision Platform: Smart Terrain for Depth-Sensing CamerasQualcomm Vuforia Mobile Vision Platform: Smart Terrain for Depth-Sensing Cameras
Qualcomm Vuforia Mobile Vision Platform: Smart Terrain for Depth-Sensing Cameras
 
Android Development Tools
Android Development ToolsAndroid Development Tools
Android Development Tools
 
GS-4136, Optimizing Game Development using AMD’s GPU PerfStudio 2, by Gordon ...
GS-4136, Optimizing Game Development using AMD’s GPU PerfStudio 2, by Gordon ...GS-4136, Optimizing Game Development using AMD’s GPU PerfStudio 2, by Gordon ...
GS-4136, Optimizing Game Development using AMD’s GPU PerfStudio 2, by Gordon ...
 
Asus Tinker Board
Asus Tinker BoardAsus Tinker Board
Asus Tinker Board
 
Webinar: NVIDIA JETSON – A Inteligência Artificial na palma de sua mão
Webinar: NVIDIA JETSON – A Inteligência Artificial na palma de sua mãoWebinar: NVIDIA JETSON – A Inteligência Artificial na palma de sua mão
Webinar: NVIDIA JETSON – A Inteligência Artificial na palma de sua mão
 
0xdroid -- community-developed Android distribution by 0xlab
0xdroid -- community-developed Android distribution by 0xlab0xdroid -- community-developed Android distribution by 0xlab
0xdroid -- community-developed Android distribution by 0xlab
 
Embedded and Reliable Computer Vision
Embedded and Reliable Computer VisionEmbedded and Reliable Computer Vision
Embedded and Reliable Computer Vision
 
Accelerate graphics performance with ozone-gbm on Intel based Linux desktop s...
Accelerate graphics performance with ozone-gbm on Intel based Linux desktop s...Accelerate graphics performance with ozone-gbm on Intel based Linux desktop s...
Accelerate graphics performance with ozone-gbm on Intel based Linux desktop s...
 
Java fx on raspberry pi gerrit grunwald
Java fx on raspberry pi   gerrit grunwaldJava fx on raspberry pi   gerrit grunwald
Java fx on raspberry pi gerrit grunwald
 
Computer Vision Powered by Heterogeneous System Architecture (HSA) by Dr. Ha...
Computer Vision Powered by Heterogeneous System Architecture (HSA) by  Dr. Ha...Computer Vision Powered by Heterogeneous System Architecture (HSA) by  Dr. Ha...
Computer Vision Powered by Heterogeneous System Architecture (HSA) by Dr. Ha...
 
Qualcomm Hexagon SDK: Optimize Your Multimedia Solutions
Qualcomm Hexagon SDK: Optimize Your Multimedia SolutionsQualcomm Hexagon SDK: Optimize Your Multimedia Solutions
Qualcomm Hexagon SDK: Optimize Your Multimedia Solutions
 
Digital Watchdog DWC-MD421DB Data Sheet
Digital Watchdog DWC-MD421DB Data SheetDigital Watchdog DWC-MD421DB Data Sheet
Digital Watchdog DWC-MD421DB Data Sheet
 
Choosing the Right IP Cores for Low-Latency Video Streaming
Choosing the Right IP Cores for Low-Latency Video StreamingChoosing the Right IP Cores for Low-Latency Video Streaming
Choosing the Right IP Cores for Low-Latency Video Streaming
 
Intel® Open Image Denoise: Optimized CPU Denoising | SIGGRAPH 2019 Technical ...
Intel® Open Image Denoise: Optimized CPU Denoising | SIGGRAPH 2019 Technical ...Intel® Open Image Denoise: Optimized CPU Denoising | SIGGRAPH 2019 Technical ...
Intel® Open Image Denoise: Optimized CPU Denoising | SIGGRAPH 2019 Technical ...
 

Similar to IoT consideration selection

Infrared simulation and processing on Nvidia platforms
Infrared simulation and processing on Nvidia platformsInfrared simulation and processing on Nvidia platforms
Infrared simulation and processing on Nvidia platformsYoss Cohen
 
Hai Tao at AI Frontiers: Deep Learning For Embedded Vision System
Hai Tao at AI Frontiers: Deep Learning For Embedded Vision SystemHai Tao at AI Frontiers: Deep Learning For Embedded Vision System
Hai Tao at AI Frontiers: Deep Learning For Embedded Vision SystemAI Frontiers
 
Linxu conj2016 96boards
Linxu conj2016 96boardsLinxu conj2016 96boards
Linxu conj2016 96boardsLF Events
 
High End Modeling & Imaging with Intel Iris Pro Graphics
High End Modeling & Imaging with Intel Iris Pro GraphicsHigh End Modeling & Imaging with Intel Iris Pro Graphics
High End Modeling & Imaging with Intel Iris Pro GraphicsIntel® Software
 
VMworld 2013: How Good is PCoIP - A Remoting Protocol Shootout
VMworld 2013: How Good is PCoIP - A Remoting Protocol ShootoutVMworld 2013: How Good is PCoIP - A Remoting Protocol Shootout
VMworld 2013: How Good is PCoIP - A Remoting Protocol ShootoutVMworld
 
Introduction to Software Defined Visualization (SDVis)
Introduction to Software Defined Visualization (SDVis)Introduction to Software Defined Visualization (SDVis)
Introduction to Software Defined Visualization (SDVis)Intel® Software
 
Scalable AI Solution cross AI platforms
Scalable AI Solution cross AI platformsScalable AI Solution cross AI platforms
Scalable AI Solution cross AI platformsKTN
 
NVIDIA vGPU - Introduction to NVIDIA Virtual GPU
NVIDIA vGPU - Introduction to NVIDIA Virtual GPUNVIDIA vGPU - Introduction to NVIDIA Virtual GPU
NVIDIA vGPU - Introduction to NVIDIA Virtual GPULee Bushen
 
How to Choose Mobile Workstation? VR Ready
How to Choose Mobile Workstation? VR ReadyHow to Choose Mobile Workstation? VR Ready
How to Choose Mobile Workstation? VR ReadyNVIDIA Taiwan
 
“Tensilica Processor Cores Enable Sensor Fusion for Robust Perception,” a Pre...
“Tensilica Processor Cores Enable Sensor Fusion for Robust Perception,” a Pre...“Tensilica Processor Cores Enable Sensor Fusion for Robust Perception,” a Pre...
“Tensilica Processor Cores Enable Sensor Fusion for Robust Perception,” a Pre...Edge AI and Vision Alliance
 
AMD Financial Analyst Day
AMD Financial Analyst DayAMD Financial Analyst Day
AMD Financial Analyst DayAMD
 
Parallel Rendering of Webpages
Parallel Rendering of WebpagesParallel Rendering of Webpages
Parallel Rendering of WebpagesLangtech
 
“Designing the Next Ultra-Low-Power Always-On Solution,” a Presentation from ...
“Designing the Next Ultra-Low-Power Always-On Solution,” a Presentation from ...“Designing the Next Ultra-Low-Power Always-On Solution,” a Presentation from ...
“Designing the Next Ultra-Low-Power Always-On Solution,” a Presentation from ...Edge AI and Vision Alliance
 
pcDuino tech talk at Carnegie Mellon University 10/14/2014
pcDuino tech talk at Carnegie Mellon University 10/14/2014pcDuino tech talk at Carnegie Mellon University 10/14/2014
pcDuino tech talk at Carnegie Mellon University 10/14/2014Jingfeng Liu
 
Hot Chips: AMD Next Gen 7nm Ryzen 4000 APU
Hot Chips: AMD Next Gen 7nm Ryzen 4000 APUHot Chips: AMD Next Gen 7nm Ryzen 4000 APU
Hot Chips: AMD Next Gen 7nm Ryzen 4000 APUAMD
 
“Jumpstart Your Edge AI Vision Application with New Development Kits from Avn...
“Jumpstart Your Edge AI Vision Application with New Development Kits from Avn...“Jumpstart Your Edge AI Vision Application with New Development Kits from Avn...
“Jumpstart Your Edge AI Vision Application with New Development Kits from Avn...Edge AI and Vision Alliance
 
2015 02 28 DotNetSpain IoT Fight
2015 02 28 DotNetSpain IoT Fight2015 02 28 DotNetSpain IoT Fight
2015 02 28 DotNetSpain IoT FightBruno Capuano
 
Gpu with cuda architecture
Gpu with cuda architectureGpu with cuda architecture
Gpu with cuda architectureDhaval Kaneria
 

Similar to IoT consideration selection (20)

Infrared simulation and processing on Nvidia platforms
Infrared simulation and processing on Nvidia platformsInfrared simulation and processing on Nvidia platforms
Infrared simulation and processing on Nvidia platforms
 
Hai Tao at AI Frontiers: Deep Learning For Embedded Vision System
Hai Tao at AI Frontiers: Deep Learning For Embedded Vision SystemHai Tao at AI Frontiers: Deep Learning For Embedded Vision System
Hai Tao at AI Frontiers: Deep Learning For Embedded Vision System
 
Linxu conj2016 96boards
Linxu conj2016 96boardsLinxu conj2016 96boards
Linxu conj2016 96boards
 
High End Modeling & Imaging with Intel Iris Pro Graphics
High End Modeling & Imaging with Intel Iris Pro GraphicsHigh End Modeling & Imaging with Intel Iris Pro Graphics
High End Modeling & Imaging with Intel Iris Pro Graphics
 
VMworld 2013: How Good is PCoIP - A Remoting Protocol Shootout
VMworld 2013: How Good is PCoIP - A Remoting Protocol ShootoutVMworld 2013: How Good is PCoIP - A Remoting Protocol Shootout
VMworld 2013: How Good is PCoIP - A Remoting Protocol Shootout
 
Introduction to Software Defined Visualization (SDVis)
Introduction to Software Defined Visualization (SDVis)Introduction to Software Defined Visualization (SDVis)
Introduction to Software Defined Visualization (SDVis)
 
Scalable AI Solution cross AI platforms
Scalable AI Solution cross AI platformsScalable AI Solution cross AI platforms
Scalable AI Solution cross AI platforms
 
Fixed-point Multi-Core DSP Application Examples
Fixed-point Multi-Core DSP Application ExamplesFixed-point Multi-Core DSP Application Examples
Fixed-point Multi-Core DSP Application Examples
 
NVIDIA vGPU - Introduction to NVIDIA Virtual GPU
NVIDIA vGPU - Introduction to NVIDIA Virtual GPUNVIDIA vGPU - Introduction to NVIDIA Virtual GPU
NVIDIA vGPU - Introduction to NVIDIA Virtual GPU
 
How to Choose Mobile Workstation? VR Ready
How to Choose Mobile Workstation? VR ReadyHow to Choose Mobile Workstation? VR Ready
How to Choose Mobile Workstation? VR Ready
 
“Tensilica Processor Cores Enable Sensor Fusion for Robust Perception,” a Pre...
“Tensilica Processor Cores Enable Sensor Fusion for Robust Perception,” a Pre...“Tensilica Processor Cores Enable Sensor Fusion for Robust Perception,” a Pre...
“Tensilica Processor Cores Enable Sensor Fusion for Robust Perception,” a Pre...
 
AMD Financial Analyst Day
AMD Financial Analyst DayAMD Financial Analyst Day
AMD Financial Analyst Day
 
Parallel Rendering of Webpages
Parallel Rendering of WebpagesParallel Rendering of Webpages
Parallel Rendering of Webpages
 
“Designing the Next Ultra-Low-Power Always-On Solution,” a Presentation from ...
“Designing the Next Ultra-Low-Power Always-On Solution,” a Presentation from ...“Designing the Next Ultra-Low-Power Always-On Solution,” a Presentation from ...
“Designing the Next Ultra-Low-Power Always-On Solution,” a Presentation from ...
 
pcDuino tech talk at Carnegie Mellon University 10/14/2014
pcDuino tech talk at Carnegie Mellon University 10/14/2014pcDuino tech talk at Carnegie Mellon University 10/14/2014
pcDuino tech talk at Carnegie Mellon University 10/14/2014
 
Hot Chips: AMD Next Gen 7nm Ryzen 4000 APU
Hot Chips: AMD Next Gen 7nm Ryzen 4000 APUHot Chips: AMD Next Gen 7nm Ryzen 4000 APU
Hot Chips: AMD Next Gen 7nm Ryzen 4000 APU
 
“Jumpstart Your Edge AI Vision Application with New Development Kits from Avn...
“Jumpstart Your Edge AI Vision Application with New Development Kits from Avn...“Jumpstart Your Edge AI Vision Application with New Development Kits from Avn...
“Jumpstart Your Edge AI Vision Application with New Development Kits from Avn...
 
Board Design and System Software
Board Design and System SoftwareBoard Design and System Software
Board Design and System Software
 
2015 02 28 DotNetSpain IoT Fight
2015 02 28 DotNetSpain IoT Fight2015 02 28 DotNetSpain IoT Fight
2015 02 28 DotNetSpain IoT Fight
 
Gpu with cuda architecture
Gpu with cuda architectureGpu with cuda architecture
Gpu with cuda architecture
 

More from Yoss Cohen

open platform for swarm training
open platform for swarm training open platform for swarm training
open platform for swarm training Yoss Cohen
 
Deep Learning - system view
Deep Learning - system viewDeep Learning - system view
Deep Learning - system viewYoss Cohen
 
Dspip deep learning syllabus
Dspip deep learning syllabusDspip deep learning syllabus
Dspip deep learning syllabusYoss Cohen
 
Autonomous car teleportation architecture
Autonomous car teleportation architectureAutonomous car teleportation architecture
Autonomous car teleportation architectureYoss Cohen
 
Motion estimation overview
Motion estimation overviewMotion estimation overview
Motion estimation overviewYoss Cohen
 
Computer Vision - Image Filters
Computer Vision - Image FiltersComputer Vision - Image Filters
Computer Vision - Image FiltersYoss Cohen
 
Intro to machine learning with scikit learn
Intro to machine learning with scikit learnIntro to machine learning with scikit learn
Intro to machine learning with scikit learnYoss Cohen
 
DASH and HTTP2.0
DASH and HTTP2.0DASH and HTTP2.0
DASH and HTTP2.0Yoss Cohen
 
HEVC Definitions and high-level syntax
HEVC Definitions and high-level syntaxHEVC Definitions and high-level syntax
HEVC Definitions and high-level syntaxYoss Cohen
 
Introduction to HEVC
Introduction to HEVCIntroduction to HEVC
Introduction to HEVCYoss Cohen
 
FFMPEG on android
FFMPEG on androidFFMPEG on android
FFMPEG on androidYoss Cohen
 
Hands-on Video Course - "RAW Video"
Hands-on Video Course - "RAW Video" Hands-on Video Course - "RAW Video"
Hands-on Video Course - "RAW Video" Yoss Cohen
 
Video quality testing
Video quality testingVideo quality testing
Video quality testingYoss Cohen
 
HEVC / H265 Hands-On course
HEVC / H265 Hands-On courseHEVC / H265 Hands-On course
HEVC / H265 Hands-On courseYoss Cohen
 
Web video standards
Web video standardsWeb video standards
Web video standardsYoss Cohen
 
Product wise computer vision development
Product wise computer vision developmentProduct wise computer vision development
Product wise computer vision developmentYoss Cohen
 
3D Video Programming for Android
3D Video Programming for Android3D Video Programming for Android
3D Video Programming for AndroidYoss Cohen
 
Video optimization for android - Part I
Video optimization for android - Part IVideo optimization for android - Part I
Video optimization for android - Part IYoss Cohen
 

More from Yoss Cohen (20)

open platform for swarm training
open platform for swarm training open platform for swarm training
open platform for swarm training
 
Deep Learning - system view
Deep Learning - system viewDeep Learning - system view
Deep Learning - system view
 
Dspip deep learning syllabus
Dspip deep learning syllabusDspip deep learning syllabus
Dspip deep learning syllabus
 
Autonomous car teleportation architecture
Autonomous car teleportation architectureAutonomous car teleportation architecture
Autonomous car teleportation architecture
 
Motion estimation overview
Motion estimation overviewMotion estimation overview
Motion estimation overview
 
Computer Vision - Image Filters
Computer Vision - Image FiltersComputer Vision - Image Filters
Computer Vision - Image Filters
 
Intro to machine learning with scikit learn
Intro to machine learning with scikit learnIntro to machine learning with scikit learn
Intro to machine learning with scikit learn
 
DASH and HTTP2.0
DASH and HTTP2.0DASH and HTTP2.0
DASH and HTTP2.0
 
HEVC Definitions and high-level syntax
HEVC Definitions and high-level syntaxHEVC Definitions and high-level syntax
HEVC Definitions and high-level syntax
 
Introduction to HEVC
Introduction to HEVCIntroduction to HEVC
Introduction to HEVC
 
FFMPEG on android
FFMPEG on androidFFMPEG on android
FFMPEG on android
 
Hands-on Video Course - "RAW Video"
Hands-on Video Course - "RAW Video" Hands-on Video Course - "RAW Video"
Hands-on Video Course - "RAW Video"
 
Video quality testing
Video quality testingVideo quality testing
Video quality testing
 
HEVC / H265 Hands-On course
HEVC / H265 Hands-On courseHEVC / H265 Hands-On course
HEVC / H265 Hands-On course
 
Web video standards
Web video standardsWeb video standards
Web video standards
 
Product wise computer vision development
Product wise computer vision developmentProduct wise computer vision development
Product wise computer vision development
 
3D Video Programming for Android
3D Video Programming for Android3D Video Programming for Android
3D Video Programming for Android
 
Analog Video
Analog Video Analog Video
Analog Video
 
WiFi Display
WiFi DisplayWiFi Display
WiFi Display
 
Video optimization for android - Part I
Video optimization for android - Part IVideo optimization for android - Part I
Video optimization for android - Part I
 

Recently uploaded

TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessPixlogix Infotech
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoffsammart93
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerThousandEyes
 
HTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesHTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesBoston Institute of Analytics
 
GenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdfGenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdflior mazor
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationMichael W. Hawkins
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...apidays
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUK Journal
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)wesley chun
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsJoaquim Jorge
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherRemote DBA Services
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...Neo4j
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProduct Anonymous
 
Developing An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of BrazilDeveloping An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of BrazilV3cube
 

Recently uploaded (20)

TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your Business
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
HTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesHTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation Strategies
 
GenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdfGenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdf
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and Myths
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
Developing An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of BrazilDeveloping An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of Brazil
 

IoT consideration selection

  • 1. Design considerations for IoT systems Yossi Cohen December 2020
  • 2. Outline • IoT Modern needs • Deep Learning networks and operations • CPU GPU TPU and NN • Considerations • Dev-boards and NPU accelerators – RPI – NVIDIA – Rockchip – Intel • Cameras 2D / 3D • Summary
  • 4. Basic Edge Processor operation • Worked as controller • Control motors / open gates • Take pictures / basic image operations • Send images to the cloud for further processing/storage/decision • Operate based on cloud decision
  • 5. IoT Modern roles • Operations done in IoT device – Read numbers, license plates, road signs – Detect faces (dogs, cats) • Count people • Segment cars, pedestrians • Scan environment / Navigate • Drive cars / Fly drones
  • 6. Neural Networks • Basic idea: Multiply each input by weights and sum • Exists for 60+ years !! – Implemented with resistors as weights • Deep Neural networks 100+ layers – Hundreds of millions or more weights – Heavy matrix operations
  • 7. LSTM/RNN • Used for – sound detectors (gun shot detector) – data series - number of clients in restaurant – Activity detector – based on accelerometers – Usually requires low processing compared to CNNs
  • 8. CNN - Classifiers • Used for – Is object X in the image – Is there a cat in the Image ? – Common implementation LaNet
  • 9. BOX Detectors • Box detectors detects where is the object located • Used to detect: – People, face, corona mask – car, road signs – cats • Known box detectors: – YOLO3 YOLO5 – SSD
  • 10. Segmenter and Multi-Instance Segmenter • Segmentor – defines class for each pixel – Pixel(X,Y) belongs to cat – FCN-8 basic implementation – Does not separate cats • Multi-instance segmentor – Separate Cats – Pixel(X,Y) belongs to cat42 – MRCNN – common implementation
  • 11. CPU and CNNs • X86, ARM – sequential instruction processing • Not suitable for matrix (DL) operations ! • Remedies • +SIMD – MMX SSE • +Multicore 2,4,8…. • Yet….
  • 12. Moore law is dead??
  • 13. GPU – more than a graphic processor • GPU are using 100s-1000s of (simple) cores • Better in parallel processing than CPU • Much better for large matrix operations 13
  • 14. GPU Examples • GPU Cards – NVIDIA 20XX – NVIDIA T4 • Can we run a GPU only IoT solution?
  • 15. TPUs • TPUs are built ONLT for deep learning • Designed for single operation layer processing • First GPUs started 4 years ago • Google Coral, Hauwei, NVIDIA, Intel Movidius, Narvana, • In Israel: Habana, Heilo • In most modern phones: Samsung apple, QCOM
  • 16. TPU All internal buses are 256B wide
  • 17. Processor selection is obvious right? • TPU is the best solution for Deep learning activities
  • 19. Edge TPU accelerators • Intel Movidius Processor – Edge AI processor / accelerator – Either chip or USB – Connects to RPI • Google TPU (Coral) – Chip or USB • Rockchip RK 1808
  • 20. Processors today are SoC • SoC which usually include: – 4-8 ARM cores – GPU (ARM Mali, NVIDIA XX, Intel XX) – Sometimes a TPU • The selection is CPU/GPU/TPU • Selection of solution is between SoC • For MVP or development phases selection is between SoC development boards !!
  • 21. Considerations • Price - $$ • Performance TFlops/Sec • Power - Watts • Longevity • Existing code base • Development boards • Connectivity • SDKs
  • 22. Price • 10-20$ Arduino • 35-75$ RPI3 / RPI4 + extended memory • 59-100$ Jetson nano 2-4GB • 100$ Intel Movidius accelerator • 115$ Google Coral dev board • 175$ Rockchip 3399 Dev board • 200$ Jetson TX2 • 400$ Jetson NX • 700$ Jetson XVG
  • 23. Performance • Performance ratio between Jetson processors
  • 24. Operating systems • Usually, ubuntu running on most devices • Android on some boards • Raspbian / Raspberry Pi OS for RPI
  • 25. Longevity • Important to check the processor lifespan • For example, in NVIDIA jetson processors
  • 26. RPI3 / 4 dev boards • 35-75$ depending on GB • Most common dev board • Control/GPIO/basic image processing • ~3-6 fps of TinyYOLO
  • 27. NVIDIA Jetson nano • 59-100$ dev board • Relatively strong GPU • 25 fps TinyYOLO Model Application Framework NVIDIA Jetson Nano Raspberry Pi 3 Tiny YOLO V3 (416×416) Object Detection Darknet 25 FPS 0.5 FPS OpenPose (256×256) Pose Estimation Caffe 14 FPS DNR VGG-19 (224×224) Classification MXNet 10 FPS 0.5 FPS Super Resolution (481×321) Image Processing PyTorch 15 FPS DNR Unet (1x512x512) Segmentation Caffe 18 FPS DNR Source: https://developer.nvidia.com/embedded/jetson-nano-dl-inference-benchmarks
  • 28. Rockchip 3399 dev board • 175 – 299$ depending on board • USB2/3 • PCIe connectivity • GPIO • TPU accelerated
  • 29. Intel Atom • UP – boards • 200-700$ depending on kit including RealSense for some options • Use OpenVino SDK for development and optimization
  • 30. Jetson NX • 399$ dev board • X4-5 stronger than the nano • Same RPI-like form factor and connectivity • emmc + PCIe • Can do:
  • 31. Software / open source • RPI has many open-source solutions • NVIDIA uses DeepStream and ISAAC SDKs to accelerate development • Intel uses OpenVino to accelerate development on its boards
  • 32. 2D Camera • MIPI / USB connectivity / HDMI • Resolution / Formats YUV, JPEG, Bayer • HDR / Pixel size • Filters, IR Cut • Lens quality, FOV, distortion • ISP quality – very important for DL • Not all cameras supported by all boards • USB camera – UVC no need for driver on Linux windows
  • 33. Placing AI in the camera • Many IoT solutions require depth camera / Lidar - Lots of processing • To reduce processor cost done inside the camera: – Intel Realsense family – Rockchip RMSL 2031000 – Veronica - Inutive
  • 34. RealSense • D455/D435/D415 cameras – IR + RGB + Projector + dual cam • Option for IMU for robots/drones • On board processor • Not fit for RPI in high-res(high data rate)
  • 35. Inutive - Veronica • Designed for robots drones AR • Low latency to avoid AR motion sickness • On chip SLAM • Built in VPU • Output depth map
  • 36. ROCKchip RMSL203 • Rockchip depth camera solution • Built in VPU/DSP • Output depthmap • Projector + IR + RGB
  • 37. Placing AI in the camera • Many IoT solutions require depth camera / Lidar - Lots of processing • To reduce processor cost done inside the camera: – Intel Realsense family – Rockchip RMSL 2031000 – Veronica - Inutive
  • 38. RealSense • D455/D435/D415 cameras – IR + RGB + Projector + dual cam • Option for IMU for robots/drones • On board processor • Not fit for RPI in high-res(high data rate)
  • 39. Inutive - Veronica • Designed for robots drones AR • Low latency to avoid AR motion sickness • On chip SLAM • Built in VPU • Output depth map
  • 40. ROCKchip RMSL203 • Rockchip depth camera solution • Built in VPU/DSP • Output depthmap • Projector + IR + RGB
  • 41. Summary • There isnt a wining solution • Robotics / drones solutions differ from video analytics • First product MVP should prefer development time & costs over price/performance • Select a product with good, well tried SDKs and community
  • 42. Cloud connectivity • Why we need it? – Heavy data/image/deep learning processing – Data aggregation from many sensing points – Logging – Dashboard – Health test – Secure access – OTA updates • Both AWS and Azure has Cloud IoT solution which seamless integrate with RPI3 and Jetson Nano
  • 43. NVIDIA and AWS IoT (Core & Greengrass) • Connecting to AWS IoT is easy • Deepstream has already built in MQTT message formatters and brokers designed for AWS connectivity Source: https://aws.amazon.com/blogs/iot/how-to-integrate-nvidia-deepstream-on-jetson-devices-with-aws-iot-core-and-aws-iot- greengrass/
  • 44. Cloud connectivity • Why we need it? – Heavy data/image/deep learning processing – Data aggregation from many sensing points – Logging – Dashboard – Health test – Secure access – OTA updates • Both AWS and Azure has Cloud IoT solution which seamless integrate with RPI3 Jetson products and Intel products
  • 45. Summary • There isn't a wining solution • Robotics / drones solutions differ from video analytics • First product MVP should prefer development time & costs over price/performance • Select a product with good, well tried SDKs and community • Invest in a good camera / ISP
  • 46. Contact • Yossi Cohen CTO@DSP-IP • +972-54-5313092 | yossi@dsp-ip.com • What we do: – Video Analytics – Robotics – Drones – Autonomous cars – Agritec

Editor's Notes

  1. We will see how the migration affects each of those steps
  2. This requires a large investment in server side software, However cloud vendors mapped all those requirements and create a generic platform which supports them
  3. This requires a large investment in server side software, However cloud vendors mapped all those requirements and create a generic platform which supports them