IVisual: An Intelligent Visual Sensor SoC with 2790fps CMOS Image Sensor and 205GOPS/W Vision Processor Video analysis technology –Healthcare, HMI, surveillance,

Slides:



Advertisements
Similar presentations
Streaming SIMD Extension (SSE)
Advertisements

EECC756 - Shaaban #1 lec # 1 Spring Systolic Architectures Replace single processor with an array of regular processing elements Orchestrate.
Borough of Verona 2014 Using axis technology. Two Types of Day and Night Vision Technologies Wide dynamic range (WDR) describes an attribute of an imaging.
Y A S O O B A L I b o r n o n 1 9 t h F e b r u a r y i n K a n p u r d i s t r i c t o f U t t a r P r a d e s h. H e s t a r t e d s i n g i.
ENERGY-PROPORTIONAL IMAGE SENSING FOR Robert LiKamWa Bodhi Priyantha Matthai Philipose Victor Bahl Lin Zhong CONTINUOUS MOBILE VISION
A 600MS/s 30mW 0.13µm CMOS ADC Array Achieving over 60dB SFDR with Adaptive Digital Equalization Time-interleaved ADC array –High sampling rate, low power.
Real-Time Video Analysis on an Embedded Smart Camera for Traffic Surveillance Presenter: Yu-Wei Fan.
Microelectronics Design Group RAL - Microelectronics Open Day 28 th June 2005 Intelligent Digital Sensors Jamie Crooks.
Power Distribution FPGADDR2 OEM Board Flash MEMSSDINSD3 Cameras Connector Board 1.2V1.8V5V3.3V 3.3V, 5V, 12V 15V3.3V9-36V.
Define Embedded Systems Small (?) Application Specific Computer Systems.
Презентація за розділом “Гумористичні твори”
Центр атестації педагогічних працівників 2014
Галактики і квазари.
Характеристика ІНДІЇ.
Процюк Н.В. вчитель початкових класів Боярської ЗОШ І – ІІІ ст №4
EEL 6935 Embedded Systems Long Presentation 2 Group Member: Qin Chen, Xiang Mao 4/2/20101.
A Time Error Correction Method Applied to High-Precision AER Asynchronous CMOS Image Sensor 李东盛.
GPGPU overview. Graphics Processing Unit (GPU) GPU is the chip in computer video cards, PS3, Xbox, etc – Designed to realize the 3D graphics pipeline.
February 12, 1998 Aman Sareen DPGA-Coupled Microprocessors Commodity IC’s for the Early 21st Century by Aman Sareen School of Electrical Engineering and.
Luis Cordero Business Development phone: mail: Spanish SME ̴ 50 employees > 90% engineers and PhD Spin-off started.
2007 DAC/ISSCC Student Design Contest Winner Jeong-Ho Woo ( ) A 152mW/195mW Multimedia Processor with Fully Programmable 3D Graphics.
Multi-core systems System Architecture COMP25212 Daniel Goodman Advanced Processor Technologies Group.
Itr3 lecture 3: the motherboard Thomas Krichel
Tracking with CACTuS on Jetson Running a Bayesian multi object tracker on a low power, embedded system School of Information Technology & Mathematical.
CS332 Visual Processing Department of Computer Science Wellesley College CS 332 Visual Processing in Computer and Biological Vision Systems Overview of.
Духовні символи Голосіївського району
ULPVIS: Ultra-Low Power Video Sensor Networks for IoT ($700B market)
CS 351/ IT 351 Modeling and Simulation Technologies HPC Architectures Dr. Jim Holten.
Copyright © 2009 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Principles of Parallel Programming First Edition by Calvin Lin Lawrence Snyder.
Parallel Processing Presented by: Wanki Ho CS147, Section 1.
Data/Frame Memory PE 0 PE 1 PE 2 PE 3 PE N … Control Instruction Memory Interconnect The SIMD Concept.
A 1.2V 26mW Configurable Multiuser Mobile MIMO-OFDM/-OFDMA Baseband Processor Motivations –Most are single user, SISO, downlink OFDM solutions –Training.
Parallel Computers Today Oak Ridge / Cray Jaguar > 1.75 PFLOPS Two Nvidia 8800 GPUs > 1 TFLOPS Intel 80- core chip > 1 TFLOPS  TFLOPS = floating.
PRESENTED BY: MOHAMAD HAMMAM ALSAFRJALANI UFL ECE Dept. 3/31/2010 UFL ECE Dept 1 CACHE OPTIMIZATION FOR AN EMBEDDED MPEG-4 VIDEO DECODER.
Efficient Opportunistic Sensing using Mobile Collaborative Platform MOSDEN.
ULTRALEV: Ultra-Low-Energy Video Sensor Networks for IoT ($700B market) Technology Research Center University of Turku Finland.
Image from
A 45nm 37.3GOPS/W Heterogeneous Multi-Core SoC ● Renesas Technology, Kodaira, Japan ● Hitachi, Kodaira, Japan ● Waseda University, Shinjuku, Japan ● Tokyo.
HUAWEI eSight: Converged Management Solution for Safe City
Microarchitecture.
Modern Data Management
The challenges of automated and connected driving
Ultra-Low-Voltage UWB Baseband Processor
SoC and FPGA Oriented High-quality Stereo Vision System
CS294-1 Reading Aug 28, 2003 Jaein Jeong
Проф. д-р Васил Цанов, Институт за икономически изследвания при БАН
ЗУТ ПРОЕКТ на Закон за изменение и допълнение на ЗУТ
О Б Щ И Н А С И Л И С Т Р А П р о е к т Б ю д ж е т г.
Електронни услуги на НАП
Боряна Георгиева – директор на
РАЙОНЕН СЪД - БУРГАС РАБОТНА СРЕЩА СЪС СЪДЕБНИТЕ ЗАСЕДАТЕЛИ ПРИ РАЙОНЕН СЪД – БУРГАС 21 ОКТОМВРИ 2016 г.
Сътрудничество между полицията и другите специалисти в България
Съобщение Ръководството на НУ “Христо Ботев“ – гр. Елин Пелин
НАЦИОНАЛНА АГЕНЦИЯ ЗА ПРИХОДИТЕ
ДОБРОВОЛЕН РЕЗЕРВ НА ВЪОРЪЖЕНИТЕ СИЛИ НА РЕПУБЛИКА БЪЛГАРИЯ
Съвременни софтуерни решения
ПО ПЧЕЛАРСТВО ЗА ТРИГОДИШНИЯ
от проучване на общественото мнение,
Васил Големански Ноември, 2006
Програма за развитие на селските райони
ОПЕРАТИВНА ПРОГРАМА “АДМИНИСТРАТИВЕН КАПАЦИТЕТ”
БАЛИСТИКА НА ТЯЛО ПРИ СВОБОДНО ПАДАНЕ В ЗЕМНАТА АТМОСФЕРА
МЕДИЦИНСКИ УНИВЕРСИТЕТ – ПЛЕВЕН
Стратегия за развитие на клъстера 2015
Моето наследствено призвание
Правна кантора “Джингов, Гугински, Кючуков & Величков”
Безопасност на движението
Overview of Computer Architecture and Organization
Introduction to Heterogeneous Parallel Computing
Complete Vehicle Systems.
Presentation transcript:

iVisual: An Intelligent Visual Sensor SoC with 2790fps CMOS Image Sensor and 205GOPS/W Vision Processor Video analysis technology –Healthcare, HMI, surveillance, intelligent vehicle… Challenges –Privacy: Video data revealed from inter-chip traffic –Throughput: Traffic mismatch slows down processors –Power: parallel access of data memory 1 XETAL-II (ISSCC’07) Data Production: 430Gb/s Data Consumption: 2.7Gb/s Parallel Memory Access (2R1W/cycle)

iVisual Architecture 2 Parallel-to-scalar processor 36% throughput increase Bitplane-based storage, reduced data port Reduce 43% process area Light-in, answer-out SoC architecture, no video data revealed to external Avoid possible privacy problems Additional memory hierarchy between SIMD array and memory 62% power reduction

Results 3 Throughput increase Power efficiency