Cascaded Stripe Memory Engines for Multi-Scale Object Detection in FPGA
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F19%3APU134943" target="_blank" >RIV/00216305:26230/19:PU134943 - isvavai.cz</a>
Výsledek na webu
<a href="https://ieeexplore.ieee.org/document/8573854" target="_blank" >https://ieeexplore.ieee.org/document/8573854</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/TCSVT.2018.2886476" target="_blank" >10.1109/TCSVT.2018.2886476</a>
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
Cascaded Stripe Memory Engines for Multi-Scale Object Detection in FPGA
Popis výsledku v původním jazyce
Object detection in embedded systems is important for many contemporary applications that involve vision and scene analysis. In this paper, we propose a novel architecture for object detection implemented in FPGA, based on the Stripe Memory Engine (SME), and point out shortcomings of existing architectures. SME processes a stream of image data so that it stores a narrow stripe of the input image and its scaled versions and uses a detector unit which is efficiently pipelined across multiple image positions within the SME. We show how to process images with up to 4K resolution at high framerates using cascades of SMEs. As a detector algorithm, the SMEs use boosted soft cascade with simple image features that require only pixel comparisons and look-up tables; therefore, they are well suitable for hardware implemenation. We describe the components of our architecture and compare it to several published works in several configurations. As an example, we implemented face detection and license plate detection applications that work with HD images (1280×720 pixels) running at over 60 frames per second on Xilinx Zynq platform. We analyzed their power consumption, evaluated the accuracy of our detectors, and compared them to Haar Cascades from OpenCV that are often used by other authors. We show that our detectors offer better accuracy as well as performance at lower power consumption.
Název v anglickém jazyce
Cascaded Stripe Memory Engines for Multi-Scale Object Detection in FPGA
Popis výsledku anglicky
Object detection in embedded systems is important for many contemporary applications that involve vision and scene analysis. In this paper, we propose a novel architecture for object detection implemented in FPGA, based on the Stripe Memory Engine (SME), and point out shortcomings of existing architectures. SME processes a stream of image data so that it stores a narrow stripe of the input image and its scaled versions and uses a detector unit which is efficiently pipelined across multiple image positions within the SME. We show how to process images with up to 4K resolution at high framerates using cascades of SMEs. As a detector algorithm, the SMEs use boosted soft cascade with simple image features that require only pixel comparisons and look-up tables; therefore, they are well suitable for hardware implemenation. We describe the components of our architecture and compare it to several published works in several configurations. As an example, we implemented face detection and license plate detection applications that work with HD images (1280×720 pixels) running at over 60 frames per second on Xilinx Zynq platform. We analyzed their power consumption, evaluated the accuracy of our detectors, and compared them to Haar Cascades from OpenCV that are often used by other authors. We show that our detectors offer better accuracy as well as performance at lower power consumption.
Klasifikace
Druh
J<sub>imp</sub> - Článek v periodiku v databázi Web of Science
CEP obor
—
OECD FORD obor
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Návaznosti výsledku
Projekt
<a href="/cs/project/LQ1602" target="_blank" >LQ1602: IT4Innovations excellence in science</a><br>
Návaznosti
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Ostatní
Rok uplatnění
2020
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Údaje specifické pro druh výsledku
Název periodika
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
ISSN
1051-8215
e-ISSN
1558-2205
Svazek periodika
30
Číslo periodika v rámci svazku
1
Stát vydavatele periodika
US - Spojené státy americké
Počet stran výsledku
13
Strana od-do
267-280
Kód UT WoS článku
000521641800022
EID výsledku v databázi Scopus
2-s2.0-85058662945