SlideShare a Scribd company logo
1 of 26
Download to read offline
© 2023 Midokura (Sony Group)
Modernizing the
Development of AI-
Based IoT Devices with
Wedge
Dan Mihai Dumitriu
CTO
Midokura (Sony Group)
© 2023 Midokura (Sony Group)
• Sony Semiconductor Solutions is the largest vendor of CMOS image sensors
• AITRIOS is a new business to develop sensing solutions based on the sensor
portfolio
• Embedded Edge AI and IoT are key ingredients
• Midokura is a subsidiary of Sony Semicon focused on software infrastructure
Introduction
2
© 2023 Midokura (Sony Group)
Sony AITRIOS
3
Address privacy concerns
by elaborating metadata-only
Low power and
cost friendly
Faster and optimized
on-sensor AI processing
Less bandwidth needed
and very low latency
AI-enabled
image sensor
(IMX500)
AI/device
management
cloud service
New type of
AI/computer vision
hardware platform
IMX500
© 2023 Midokura (Sony Group)
An Accessible Platform that Enables Intelligent
Solutions Based on Distributed Visual Sensing
4
Advanced vehicle
assistance
Smart
Manufacturing
Retail stores
Retail stores
Smart Home
Bring to market a
plethora of low cost,
easy-to-use sensing
devices
Low barrier of entry for
solution developers
Agile development of
sensing applications
Low operational cost of
vision sensing apps
Polyglot Development Marketplace to connect
AI Developers & Solution
Developers
Targeting solution developers for various vertical applications.
© 2023 Midokura (Sony Group) 5
Problems of Development on Tiny
IoT Devices
© 2023 Midokura (Sony Group)
• Operating system (RTOS) is effectively like a library
• Application and OS must be tested together
• Waterfall development
• Heavy testing process late in development cycle
• Cannot continuously deploy applications
Development is Not Agile
6
© 2023 Midokura (Sony Group)
• Tiny IoT devices are based on MCUs (microcontroller units)
• MCU typically cannot do virtual memory
• Without memory isolation, cannot do agile development
• Dynamic deployment of applications is unsafe
No Safety and Isolation
7
© 2023 Midokura (Sony Group)
• High barrier of entry for application developers
• Apps are typically written in C
• Poor code reuse across device types
• HW specific interfaces and drivers
• AI developers typically don't know C well
• Develop mostly in Python
Application Development Is Difficult
8
© 2023 Midokura (Sony Group) 9
Introducing WEdge
© 2023 Midokura (Sony Group)
• Like Kubernetes, but for tiny IoT devices
• Automated lifecycle management of workloads
• Strong isolation of modules
• Leverages WebAssembly Micro Runtime (WAMR)
• WEdge cloud does automatic optimization for target device
• Polyglot SDK enables developer productivity
Intro to WEdge
10
© 2023 Midokura (Sony Group)
• WebAssembly (Wasm) is a low-level bytecode format that runs in a sandboxed
environment
• Designed as a portable target for compiling high-level languages like C, C++, and Rust
• Runs on various platforms, including browsers, servers, and now even on tiny IoT devices
• Compact size, fast execution, and high security compared to other runtime environments
• AoT (Ahead of Time) compilation supports multiple target ISAs via LLVM backends
• High runtime performance - within 2x of native
Intro to WebAssembly
11
© 2023 Midokura (Sony Group)
• Combination of language-level and runtime-level protections
• Linear memory model
• All memory accesses are bounds-checked to prevent buffer overflows or
underflows
• Enforces type safety at both compile-time and runtime
• Control Flow Integrity (CFI) prevents control flow hijacking attacks
• WebAssembly System Interface (WASI) for standardized and secure way to
access system resources
• Fine grained control
WebAssembly for Sandboxing
12
© 2023 Midokura (Sony Group)
WEdge Device Stack
13
WEdge Services API
OS
Abstraction
Layer
WASI
Web Assembly Micro Runtime
Native Libraries & Device Drivers
OS + BSP
HW
Module
1
. . .
Module
2
Module
3
Module
4
Module
N
© 2023 Midokura (Sony Group)
• Sensors
• Read image
• Configure (e.g. frame rate)
• Communication
• Send telemetry to cloud
• HTTP PUT/GET/POST
• Other protocols (e.g. CAN)
WEdge Services API
14
• AI/ML
• Load model
• Run inference
•
Data storage
• Local DB
• Distributed DB cache
© 2023 Midokura (Sony Group)
SDKs in various languages
Polyglot Development
15
© 2023 Midokura (Sony Group)
Decouple from Target Architecture & OS
16
© 2023 Midokura (Sony Group)
• Enables more flexible and dynamic development cycles
• Devs can break down complex software into smaller, more
manageable modules that can be developed, tested, and
deployed independently
• Polyglot development helps devs code in their language of choice
• For AI developers, that is Python
Agile Development Enabler
17
© 2023 Midokura (Sony Group) 18
Vision Sensing Applications
© 2023 Midokura (Sony Group)
• Programming sensing applications is complex
• Code reuse is typically poor and applications are monolithic
• VSA models a complex application as a series of simple nodes
• Promotes code reuse
• Enables low-code development via API or GUI
• VSA nodes are deployed as Wasm modules
Vision Sensing Applications (VSA)
19
© 2023 Midokura (Sony Group)
Goal: Make an intelligent sensor perform a complex task
How: Model this complex task as a sequence of individual small tasks
Complex task => VSA
Small tasks (partial) => Nodes of the VSA
Why?
● Reusable nodes - can have different creators
● Allow to customize an existing application to a new use case by modifying some nodes
● Security: Use Webassembly technology with its sandboxing approach to allow custom
code without impact on the device security
Vision Sensing Application (VSA)
20
© 2023 Midokura (Sony Group) 21
Example Application: People Counting
© 2023 Midokura (Sony Group)
People Counting VSA
22
© 2023 Midokura (Sony Group)
VSA of People Counting Application
23
© 2023 Midokura (Sony Group)
• WebAssembly is a great choice for tiny IoT devices
• High performance and low overhead
• WEdge enables high development productivity and agility
• Vision Sensing Applications (VSAs) are a convenient paradigm for
programming sensing applications
• Promoting component reuse
• Enabling low-/no-code development
Conclusion
24
© 2023 Midokura (Sony Group)
Further Information
25
Midokura
https://www.midokura.com
AITRIOS
https://www.aitrios.sony-semicon.com/en/
IMX500
https://developer.sony.com/develop/imx500/
Demo Video
https://bit.ly/mido-vsa-demo
Please visit our booth!
© 2023 Midokura (Sony Group) 26
Thank you

More Related Content

What's hot

김혁, <드래곤 하운드>의 PBR과 레이트레이싱 렌더링 기법, NDC2019
김혁, <드래곤 하운드>의 PBR과 레이트레이싱 렌더링 기법, NDC2019김혁, <드래곤 하운드>의 PBR과 레이트레이싱 렌더링 기법, NDC2019
김혁, <드래곤 하운드>의 PBR과 레이트레이싱 렌더링 기법, NDC2019devCAT Studio, NEXON
 
Pose estimation from RGB images by deep learning
Pose estimation from RGB images by deep learningPose estimation from RGB images by deep learning
Pose estimation from RGB images by deep learningYu Huang
 
Deferred decal
Deferred decalDeferred decal
Deferred decal민웅 이
 
Deep learning - A Visual Introduction
Deep learning - A Visual IntroductionDeep learning - A Visual Introduction
Deep learning - A Visual IntroductionLukas Masuch
 
Intro to Deep Learning for Computer Vision
Intro to Deep Learning for Computer VisionIntro to Deep Learning for Computer Vision
Intro to Deep Learning for Computer VisionChristoph Körner
 
Deep neural networks
Deep neural networksDeep neural networks
Deep neural networksSi Haem
 
MIPI DevCon 2016: MIPI CSI-2 Application for Vision and Sensor Fusion Systems
MIPI DevCon 2016: MIPI CSI-2 Application for Vision and Sensor Fusion SystemsMIPI DevCon 2016: MIPI CSI-2 Application for Vision and Sensor Fusion Systems
MIPI DevCon 2016: MIPI CSI-2 Application for Vision and Sensor Fusion SystemsMIPI Alliance
 
Latent diffusions vs DALL-E v2
Latent diffusions vs DALL-E v2Latent diffusions vs DALL-E v2
Latent diffusions vs DALL-E v2Vitaly Bondar
 
Convolutional Neural Networks
Convolutional Neural NetworksConvolutional Neural Networks
Convolutional Neural NetworksAshray Bhandare
 
Human Pose Estimation by Deep Learning
Human Pose Estimation by Deep LearningHuman Pose Estimation by Deep Learning
Human Pose Estimation by Deep LearningWei Yang
 
Simple Introduction to AutoEncoder
Simple Introduction to AutoEncoderSimple Introduction to AutoEncoder
Simple Introduction to AutoEncoderJun Lang
 
Computer vision and Open CV
Computer vision and Open CVComputer vision and Open CV
Computer vision and Open CVChariza Pladin
 
MIPI DevCon 2021: MIPI CSI-2 v4.0 Panel Discussion with the MIPI Camera Worki...
MIPI DevCon 2021: MIPI CSI-2 v4.0 Panel Discussion with the MIPI Camera Worki...MIPI DevCon 2021: MIPI CSI-2 v4.0 Panel Discussion with the MIPI Camera Worki...
MIPI DevCon 2021: MIPI CSI-2 v4.0 Panel Discussion with the MIPI Camera Worki...MIPI Alliance
 
Feature Extraction
Feature ExtractionFeature Extraction
Feature Extractionskylian
 
Guided image filter
Guided image filterGuided image filter
Guided image filterssuser456ad6
 

What's hot (20)

김혁, <드래곤 하운드>의 PBR과 레이트레이싱 렌더링 기법, NDC2019
김혁, <드래곤 하운드>의 PBR과 레이트레이싱 렌더링 기법, NDC2019김혁, <드래곤 하운드>의 PBR과 레이트레이싱 렌더링 기법, NDC2019
김혁, <드래곤 하운드>의 PBR과 레이트레이싱 렌더링 기법, NDC2019
 
Pose estimation from RGB images by deep learning
Pose estimation from RGB images by deep learningPose estimation from RGB images by deep learning
Pose estimation from RGB images by deep learning
 
Deferred decal
Deferred decalDeferred decal
Deferred decal
 
Deep learning - A Visual Introduction
Deep learning - A Visual IntroductionDeep learning - A Visual Introduction
Deep learning - A Visual Introduction
 
Intro to Deep Learning for Computer Vision
Intro to Deep Learning for Computer VisionIntro to Deep Learning for Computer Vision
Intro to Deep Learning for Computer Vision
 
Deep neural networks
Deep neural networksDeep neural networks
Deep neural networks
 
WBF.pptx
WBF.pptxWBF.pptx
WBF.pptx
 
MIPI DevCon 2016: MIPI CSI-2 Application for Vision and Sensor Fusion Systems
MIPI DevCon 2016: MIPI CSI-2 Application for Vision and Sensor Fusion SystemsMIPI DevCon 2016: MIPI CSI-2 Application for Vision and Sensor Fusion Systems
MIPI DevCon 2016: MIPI CSI-2 Application for Vision and Sensor Fusion Systems
 
Latent diffusions vs DALL-E v2
Latent diffusions vs DALL-E v2Latent diffusions vs DALL-E v2
Latent diffusions vs DALL-E v2
 
U-Net (1).pptx
U-Net (1).pptxU-Net (1).pptx
U-Net (1).pptx
 
Convolutional Neural Networks
Convolutional Neural NetworksConvolutional Neural Networks
Convolutional Neural Networks
 
Computer vision
Computer visionComputer vision
Computer vision
 
Human Pose Estimation by Deep Learning
Human Pose Estimation by Deep LearningHuman Pose Estimation by Deep Learning
Human Pose Estimation by Deep Learning
 
Image captioning
Image captioningImage captioning
Image captioning
 
Federated Learning
Federated LearningFederated Learning
Federated Learning
 
Simple Introduction to AutoEncoder
Simple Introduction to AutoEncoderSimple Introduction to AutoEncoder
Simple Introduction to AutoEncoder
 
Computer vision and Open CV
Computer vision and Open CVComputer vision and Open CV
Computer vision and Open CV
 
MIPI DevCon 2021: MIPI CSI-2 v4.0 Panel Discussion with the MIPI Camera Worki...
MIPI DevCon 2021: MIPI CSI-2 v4.0 Panel Discussion with the MIPI Camera Worki...MIPI DevCon 2021: MIPI CSI-2 v4.0 Panel Discussion with the MIPI Camera Worki...
MIPI DevCon 2021: MIPI CSI-2 v4.0 Panel Discussion with the MIPI Camera Worki...
 
Feature Extraction
Feature ExtractionFeature Extraction
Feature Extraction
 
Guided image filter
Guided image filterGuided image filter
Guided image filter
 

Similar to “Modernizing the Development of AI-based IoT Devices with Wedge,” a Presentation from Midokura, a Sony Group Company

“Sparking the Next Generation of Arm-based Cloud-native Smart Camera Designs,...
“Sparking the Next Generation of Arm-based Cloud-native Smart Camera Designs,...“Sparking the Next Generation of Arm-based Cloud-native Smart Camera Designs,...
“Sparking the Next Generation of Arm-based Cloud-native Smart Camera Designs,...Edge AI and Vision Alliance
 
Cloudmov: Cloud-Based Mobile Social TV
Cloudmov: Cloud-Based Mobile Social TVCloudmov: Cloud-Based Mobile Social TV
Cloudmov: Cloud-Based Mobile Social TVIRJET Journal
 
“Building Large-scale Distributed Computer Vision Solutions Without Starting ...
“Building Large-scale Distributed Computer Vision Solutions Without Starting ...“Building Large-scale Distributed Computer Vision Solutions Without Starting ...
“Building Large-scale Distributed Computer Vision Solutions Without Starting ...Edge AI and Vision Alliance
 
UNICORE Project Technical Overview
UNICORE Project Technical OverviewUNICORE Project Technical Overview
UNICORE Project Technical OverviewUNICORE_project
 
Industrial IoT Mayhem? Java IoT Gateways to the Rescue
Industrial IoT Mayhem? Java IoT Gateways to the RescueIndustrial IoT Mayhem? Java IoT Gateways to the Rescue
Industrial IoT Mayhem? Java IoT Gateways to the RescueEurotech
 
IoTSummit: Create iot devices connected or on the edge using ai and ml
IoTSummit: Create iot devices connected or on the edge using ai and mlIoTSummit: Create iot devices connected or on the edge using ai and ml
IoTSummit: Create iot devices connected or on the edge using ai and mlMarco Dal Pino
 
Software virtualization lessons for extreme IoT portability and scale
Software virtualization lessons for extreme IoT portability and scaleSoftware virtualization lessons for extreme IoT portability and scale
Software virtualization lessons for extreme IoT portability and scaleMicroEJ
 
Connect a chips to Azure
Connect a chips to AzureConnect a chips to Azure
Connect a chips to AzureMirco Vanini
 
VF2018: 'mimik' the cloud to the edge (Fay Arjomandi, Founder and CPO, mimik)
VF2018:  'mimik' the cloud to the edge (Fay Arjomandi, Founder and CPO, mimik)VF2018:  'mimik' the cloud to the edge (Fay Arjomandi, Founder and CPO, mimik)
VF2018: 'mimik' the cloud to the edge (Fay Arjomandi, Founder and CPO, mimik)Craig Asano
 
IoTSummit: Design and architect always disconnected iot system
IoTSummit: Design and architect always disconnected iot systemIoTSummit: Design and architect always disconnected iot system
IoTSummit: Design and architect always disconnected iot systemMarco Dal Pino
 
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...Edge AI and Vision Alliance
 
Kura M2M IoT Gateway
Kura M2M IoT GatewayKura M2M IoT Gateway
Kura M2M IoT GatewayEurotech
 
Build an azure connected io t device in 45 minutes (or less)
Build an azure connected io t device in 45 minutes (or less)Build an azure connected io t device in 45 minutes (or less)
Build an azure connected io t device in 45 minutes (or less)Marco Dal Pino
 

Similar to “Modernizing the Development of AI-based IoT Devices with Wedge,” a Presentation from Midokura, a Sony Group Company (20)

“Sparking the Next Generation of Arm-based Cloud-native Smart Camera Designs,...
“Sparking the Next Generation of Arm-based Cloud-native Smart Camera Designs,...“Sparking the Next Generation of Arm-based Cloud-native Smart Camera Designs,...
“Sparking the Next Generation of Arm-based Cloud-native Smart Camera Designs,...
 
Midori ppt
Midori pptMidori ppt
Midori ppt
 
Cloudmov: Cloud-Based Mobile Social TV
Cloudmov: Cloud-Based Mobile Social TVCloudmov: Cloud-Based Mobile Social TV
Cloudmov: Cloud-Based Mobile Social TV
 
“Building Large-scale Distributed Computer Vision Solutions Without Starting ...
“Building Large-scale Distributed Computer Vision Solutions Without Starting ...“Building Large-scale Distributed Computer Vision Solutions Without Starting ...
“Building Large-scale Distributed Computer Vision Solutions Without Starting ...
 
UNICORE Project Technical Overview
UNICORE Project Technical OverviewUNICORE Project Technical Overview
UNICORE Project Technical Overview
 
Industrial IoT Mayhem? Java IoT Gateways to the Rescue
Industrial IoT Mayhem? Java IoT Gateways to the RescueIndustrial IoT Mayhem? Java IoT Gateways to the Rescue
Industrial IoT Mayhem? Java IoT Gateways to the Rescue
 
Windows for IOT
Windows for IOTWindows for IOT
Windows for IOT
 
IoTSummit: Create iot devices connected or on the edge using ai and ml
IoTSummit: Create iot devices connected or on the edge using ai and mlIoTSummit: Create iot devices connected or on the edge using ai and ml
IoTSummit: Create iot devices connected or on the edge using ai and ml
 
Software virtualization lessons for extreme IoT portability and scale
Software virtualization lessons for extreme IoT portability and scaleSoftware virtualization lessons for extreme IoT portability and scale
Software virtualization lessons for extreme IoT portability and scale
 
Connect a chips to Azure
Connect a chips to AzureConnect a chips to Azure
Connect a chips to Azure
 
VF2018: 'mimik' the cloud to the edge (Fay Arjomandi, Founder and CPO, mimik)
VF2018:  'mimik' the cloud to the edge (Fay Arjomandi, Founder and CPO, mimik)VF2018:  'mimik' the cloud to the edge (Fay Arjomandi, Founder and CPO, mimik)
VF2018: 'mimik' the cloud to the edge (Fay Arjomandi, Founder and CPO, mimik)
 
Symbian
SymbianSymbian
Symbian
 
Symbian
SymbianSymbian
Symbian
 
Symbian os
Symbian osSymbian os
Symbian os
 
IoTSummit: Design and architect always disconnected iot system
IoTSummit: Design and architect always disconnected iot systemIoTSummit: Design and architect always disconnected iot system
IoTSummit: Design and architect always disconnected iot system
 
Introduction to Android.ppt
Introduction to Android.pptIntroduction to Android.ppt
Introduction to Android.ppt
 
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
“The Future of AI is Here Today: Deep Dive into Qualcomm’s On-Device AI Offer...
 
Project
ProjectProject
Project
 
Kura M2M IoT Gateway
Kura M2M IoT GatewayKura M2M IoT Gateway
Kura M2M IoT Gateway
 
Build an azure connected io t device in 45 minutes (or less)
Build an azure connected io t device in 45 minutes (or less)Build an azure connected io t device in 45 minutes (or less)
Build an azure connected io t device in 45 minutes (or less)
 

More from Edge AI and Vision Alliance

“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...Edge AI and Vision Alliance
 
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...Edge AI and Vision Alliance
 
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...Edge AI and Vision Alliance
 
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...Edge AI and Vision Alliance
 
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...Edge AI and Vision Alliance
 
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...Edge AI and Vision Alliance
 
“Vision-language Representations for Robotics,” a Presentation from the Unive...
“Vision-language Representations for Robotics,” a Presentation from the Unive...“Vision-language Representations for Robotics,” a Presentation from the Unive...
“Vision-language Representations for Robotics,” a Presentation from the Unive...Edge AI and Vision Alliance
 
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsightsEdge AI and Vision Alliance
 
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...Edge AI and Vision Alliance
 
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...Edge AI and Vision Alliance
 
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...Edge AI and Vision Alliance
 
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...Edge AI and Vision Alliance
 
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...Edge AI and Vision Alliance
 
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...Edge AI and Vision Alliance
 
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...Edge AI and Vision Alliance
 
“Updating the Edge ML Development Process,” a Presentation from Samsara
“Updating the Edge ML Development Process,” a Presentation from Samsara“Updating the Edge ML Development Process,” a Presentation from Samsara
“Updating the Edge ML Development Process,” a Presentation from SamsaraEdge AI and Vision Alliance
 
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...“Combating Bias in Production Computer Vision Systems,” a Presentation from R...
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...Edge AI and Vision Alliance
 
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...Edge AI and Vision Alliance
 
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...Edge AI and Vision Alliance
 
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...Edge AI and Vision Alliance
 

More from Edge AI and Vision Alliance (20)

“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
“Learning Compact DNN Models for Embedded Vision,” a Presentation from the Un...
 
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...
“Introduction to Computer Vision with CNNs,” a Presentation from Mohammad Hag...
 
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...
“Selecting Tools for Developing, Monitoring and Maintaining ML Models,” a Pre...
 
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...
“Building Accelerated GStreamer Applications for Video and Audio AI,” a Prese...
 
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...
“Understanding, Selecting and Optimizing Object Detectors for Edge Applicatio...
 
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...
“Introduction to Modern LiDAR for Machine Perception,” a Presentation from th...
 
“Vision-language Representations for Robotics,” a Presentation from the Unive...
“Vision-language Representations for Robotics,” a Presentation from the Unive...“Vision-language Representations for Robotics,” a Presentation from the Unive...
“Vision-language Representations for Robotics,” a Presentation from the Unive...
 
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights
“ADAS and AV Sensors: What’s Winning and Why?,” a Presentation from TechInsights
 
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...
“Computer Vision in Sports: Scalable Solutions for Downmarkets,” a Presentati...
 
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...
“Detecting Data Drift in Image Classification Neural Networks,” a Presentatio...
 
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
“Deep Neural Network Training: Diagnosing Problems and Implementing Solutions...
 
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...
“AI Start-ups: The Perils of Fishing for Whales (War Stories from the Entrepr...
 
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...
“A Computer Vision System for Autonomous Satellite Maneuvering,” a Presentati...
 
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...
“Bias in Computer Vision—It’s Bigger Than Facial Recognition!,” a Presentatio...
 
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...
“Sensor Fusion Techniques for Accurate Perception of Objects in the Environme...
 
“Updating the Edge ML Development Process,” a Presentation from Samsara
“Updating the Edge ML Development Process,” a Presentation from Samsara“Updating the Edge ML Development Process,” a Presentation from Samsara
“Updating the Edge ML Development Process,” a Presentation from Samsara
 
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...“Combating Bias in Production Computer Vision Systems,” a Presentation from R...
“Combating Bias in Production Computer Vision Systems,” a Presentation from R...
 
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...
“Developing an Embedded Vision AI-powered Fitness System,” a Presentation fro...
 
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...
“Navigating the Evolving Venture Capital Landscape for Edge AI Start-ups,” a ...
 
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...
“Advanced Presence Sensing: What It Means for the Smart Home,” a Presentation...
 

Recently uploaded

Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot ModelMcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot ModelDeepika Singh
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxRustici Software
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Jeffrey Haguewood
 
JohnPollard-hybrid-app-RailsConf2024.pptx
JohnPollard-hybrid-app-RailsConf2024.pptxJohnPollard-hybrid-app-RailsConf2024.pptx
JohnPollard-hybrid-app-RailsConf2024.pptxJohnPollard37
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...DianaGray10
 
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Zilliz
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FMESafe Software
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDropbox
 
Six Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal OntologySix Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal Ontologyjohnbeverley2021
 
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
 
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusExploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusZilliz
 
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Victor Rentea
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...Zilliz
 
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamDEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamUiPathCommunity
 
Simplifying Mobile A11y Presentation.pptx
Simplifying Mobile A11y Presentation.pptxSimplifying Mobile A11y Presentation.pptx
Simplifying Mobile A11y Presentation.pptxMarkSteadman7
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businesspanagenda
 
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfOrbitshub
 
AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)
AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)
AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)Samir Dash
 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Victor Rentea
 

Recently uploaded (20)

Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot ModelMcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptx
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
 
JohnPollard-hybrid-app-RailsConf2024.pptx
JohnPollard-hybrid-app-RailsConf2024.pptxJohnPollard-hybrid-app-RailsConf2024.pptx
JohnPollard-hybrid-app-RailsConf2024.pptx
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
 
Six Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal OntologySix Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal Ontology
 
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
 
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusExploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with Milvus
 
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamDEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
 
+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...
 
Simplifying Mobile A11y Presentation.pptx
Simplifying Mobile A11y Presentation.pptxSimplifying Mobile A11y Presentation.pptx
Simplifying Mobile A11y Presentation.pptx
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
 
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
 
AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)
AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)
AI+A11Y 11MAY2024 HYDERBAD GAAD 2024 - HelloA11Y (11 May 2024)
 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
 

“Modernizing the Development of AI-based IoT Devices with Wedge,” a Presentation from Midokura, a Sony Group Company

  • 1. © 2023 Midokura (Sony Group) Modernizing the Development of AI- Based IoT Devices with Wedge Dan Mihai Dumitriu CTO Midokura (Sony Group)
  • 2. © 2023 Midokura (Sony Group) • Sony Semiconductor Solutions is the largest vendor of CMOS image sensors • AITRIOS is a new business to develop sensing solutions based on the sensor portfolio • Embedded Edge AI and IoT are key ingredients • Midokura is a subsidiary of Sony Semicon focused on software infrastructure Introduction 2
  • 3. © 2023 Midokura (Sony Group) Sony AITRIOS 3 Address privacy concerns by elaborating metadata-only Low power and cost friendly Faster and optimized on-sensor AI processing Less bandwidth needed and very low latency AI-enabled image sensor (IMX500) AI/device management cloud service New type of AI/computer vision hardware platform IMX500
  • 4. © 2023 Midokura (Sony Group) An Accessible Platform that Enables Intelligent Solutions Based on Distributed Visual Sensing 4 Advanced vehicle assistance Smart Manufacturing Retail stores Retail stores Smart Home Bring to market a plethora of low cost, easy-to-use sensing devices Low barrier of entry for solution developers Agile development of sensing applications Low operational cost of vision sensing apps Polyglot Development Marketplace to connect AI Developers & Solution Developers Targeting solution developers for various vertical applications.
  • 5. © 2023 Midokura (Sony Group) 5 Problems of Development on Tiny IoT Devices
  • 6. © 2023 Midokura (Sony Group) • Operating system (RTOS) is effectively like a library • Application and OS must be tested together • Waterfall development • Heavy testing process late in development cycle • Cannot continuously deploy applications Development is Not Agile 6
  • 7. © 2023 Midokura (Sony Group) • Tiny IoT devices are based on MCUs (microcontroller units) • MCU typically cannot do virtual memory • Without memory isolation, cannot do agile development • Dynamic deployment of applications is unsafe No Safety and Isolation 7
  • 8. © 2023 Midokura (Sony Group) • High barrier of entry for application developers • Apps are typically written in C • Poor code reuse across device types • HW specific interfaces and drivers • AI developers typically don't know C well • Develop mostly in Python Application Development Is Difficult 8
  • 9. © 2023 Midokura (Sony Group) 9 Introducing WEdge
  • 10. © 2023 Midokura (Sony Group) • Like Kubernetes, but for tiny IoT devices • Automated lifecycle management of workloads • Strong isolation of modules • Leverages WebAssembly Micro Runtime (WAMR) • WEdge cloud does automatic optimization for target device • Polyglot SDK enables developer productivity Intro to WEdge 10
  • 11. © 2023 Midokura (Sony Group) • WebAssembly (Wasm) is a low-level bytecode format that runs in a sandboxed environment • Designed as a portable target for compiling high-level languages like C, C++, and Rust • Runs on various platforms, including browsers, servers, and now even on tiny IoT devices • Compact size, fast execution, and high security compared to other runtime environments • AoT (Ahead of Time) compilation supports multiple target ISAs via LLVM backends • High runtime performance - within 2x of native Intro to WebAssembly 11
  • 12. © 2023 Midokura (Sony Group) • Combination of language-level and runtime-level protections • Linear memory model • All memory accesses are bounds-checked to prevent buffer overflows or underflows • Enforces type safety at both compile-time and runtime • Control Flow Integrity (CFI) prevents control flow hijacking attacks • WebAssembly System Interface (WASI) for standardized and secure way to access system resources • Fine grained control WebAssembly for Sandboxing 12
  • 13. © 2023 Midokura (Sony Group) WEdge Device Stack 13 WEdge Services API OS Abstraction Layer WASI Web Assembly Micro Runtime Native Libraries & Device Drivers OS + BSP HW Module 1 . . . Module 2 Module 3 Module 4 Module N
  • 14. © 2023 Midokura (Sony Group) • Sensors • Read image • Configure (e.g. frame rate) • Communication • Send telemetry to cloud • HTTP PUT/GET/POST • Other protocols (e.g. CAN) WEdge Services API 14 • AI/ML • Load model • Run inference • Data storage • Local DB • Distributed DB cache
  • 15. © 2023 Midokura (Sony Group) SDKs in various languages Polyglot Development 15
  • 16. © 2023 Midokura (Sony Group) Decouple from Target Architecture & OS 16
  • 17. © 2023 Midokura (Sony Group) • Enables more flexible and dynamic development cycles • Devs can break down complex software into smaller, more manageable modules that can be developed, tested, and deployed independently • Polyglot development helps devs code in their language of choice • For AI developers, that is Python Agile Development Enabler 17
  • 18. © 2023 Midokura (Sony Group) 18 Vision Sensing Applications
  • 19. © 2023 Midokura (Sony Group) • Programming sensing applications is complex • Code reuse is typically poor and applications are monolithic • VSA models a complex application as a series of simple nodes • Promotes code reuse • Enables low-code development via API or GUI • VSA nodes are deployed as Wasm modules Vision Sensing Applications (VSA) 19
  • 20. © 2023 Midokura (Sony Group) Goal: Make an intelligent sensor perform a complex task How: Model this complex task as a sequence of individual small tasks Complex task => VSA Small tasks (partial) => Nodes of the VSA Why? ● Reusable nodes - can have different creators ● Allow to customize an existing application to a new use case by modifying some nodes ● Security: Use Webassembly technology with its sandboxing approach to allow custom code without impact on the device security Vision Sensing Application (VSA) 20
  • 21. © 2023 Midokura (Sony Group) 21 Example Application: People Counting
  • 22. © 2023 Midokura (Sony Group) People Counting VSA 22
  • 23. © 2023 Midokura (Sony Group) VSA of People Counting Application 23
  • 24. © 2023 Midokura (Sony Group) • WebAssembly is a great choice for tiny IoT devices • High performance and low overhead • WEdge enables high development productivity and agility • Vision Sensing Applications (VSAs) are a convenient paradigm for programming sensing applications • Promoting component reuse • Enabling low-/no-code development Conclusion 24
  • 25. © 2023 Midokura (Sony Group) Further Information 25 Midokura https://www.midokura.com AITRIOS https://www.aitrios.sony-semicon.com/en/ IMX500 https://developer.sony.com/develop/imx500/ Demo Video https://bit.ly/mido-vsa-demo Please visit our booth!
  • 26. © 2023 Midokura (Sony Group) 26 Thank you