Time-Shared Execution of Realtime Computer Vision Pipelines by Dynamic Partial Reconfiguration

05/26/2018
by   Marie Nguyen, et al.
0

This paper presents an FPGA runtime framework that demonstrates the feasibility of using dynamic partial reconfiguration (DPR) for time-sharing an FPGA by multiple realtime computer vision pipelines. The presented time-sharing runtime framework manages an FPGA fabric that can be round-robin time-shared by different pipelines at the time scale of individual frames. In this new use-case, the challenge is to achieve useful performance despite high reconfiguration time. The paper describes the basic runtime support as well as four optimizations necessary to achieve realtime performance given the limitations of DPR on today's FPGAs. The paper provides a characterization of a working runtime framework prototype on a Xilinx ZC706 development board. The paper also reports the performance of realtime computer vision pipelines when time-shared.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/26/2018

Time-Shared Execution of Realtime Streaming Pipelines by Dynamic Partial Reconfiguration

This paper presents an FPGA runtime framework that demonstrates the feas...
research
10/23/2017

Amorphous Dynamic Partial Reconfiguration with Flexible Boundaries to Remove Fragmentation

Dynamic partial reconfiguration (DPR) allows one region of an field-prog...
research
06/07/2023

BU-CVKit: Extendable Computer Vision Framework for Species Independent Tracking and Analysis

A major bottleneck of interdisciplinary computer vision (CV) research is...
research
07/31/2020

Partial Reconfiguration for Design Optimization

FPGA designers have traditionally shared a similar design methodology wi...
research
02/05/2020

MKPipe: A Compiler Framework for Optimizing Multi-Kernel Workloads in OpenCL for FPGA

OpenCL for FPGA enables developers to design FPGAs using a programming m...
research
07/05/2021

Exploring Data Pipelines through the Process Lens: a Reference Model forComputer Vision

Researchers have identified datasets used for training computer vision (...
research
09/07/2022

Auto-TransRL: Autonomous Composition of Vision Pipelines for Robotic Perception

Creating a vision pipeline for different datasets to solve a computer vi...

Please sign up or login with your details

Forgot password? Click here to reset