Corpus ID: 61282410

Automated Design of Application-Specific Smart Camera Architectures

@inproceedings{Caarls2008AutomatedDO,
  title={Automated Design of Application-Specific Smart Camera Architectures},
  author={W. Caarls},
  year={2008}
}
Parallel heterogeneous multiprocessor systems are often shunned in embedded system design, not only because of their design complexity but because of the programming burden. Programs for such systems are architecture-dependent: the application developer needs architecture-specific knowledge to implement his algorithms, as each processor has its own characteristics and programming language. He will therefore often stick to the architectures he knows best instead of looking for the best one. This… Expand
Skeleton-based automatic parallelization of image processing algorithms for GPUs
TLDR
This paper presents a technique to automatically parallelize and map sequential code on a GPU, without the need for code-annotations, and uses domain specific skeletons and a finer-grained classification of algorithms. Expand
Saliency Detection on FPGA Using Accelerators and Evaluation of Algorithmic Skeletons
Real-time vision applications are becoming more interesting to use as more computing power is available, but often those applications are still too compute intensive. Therefore, an FPGA can be usedExpand
Processor architecture design for smart cameras
TLDR
This thesis introduces a new type of SIMD architecture, called RC-SIMD, with a run-time reconfigurable communication network, and proposes a framework based on algorithmic skeletons for SmartCam solutions, which shows that variable distance communication can be achieved at a reasonable cost and substantial performance improvement. Expand
High level synthesis:performance analysis and code optimization
TLDR
The benchmarks using two image processing kernels have shown that it is possible to efficiently mimic a datapath from a reference manual RTL design and it is shown that small modifications at behavioral level can have a large positive impact on the performance results without requiring time-consuming and error-prone RTL modifications. Expand
Embedded platform selection based on the Roofline model
TLDR
The method focusses on throughput oriented systems and extends the Roofline model to heterogeneous platforms, in order to give an upper bound for the performance of an application on a platform while providing an insightful visualization of the attainable performance. Expand
Accelerating Sequential Computer Vision Algorithms Using Commodity Parallel Hardware
The last decade has seen an increasing demand from the industrial field of computerized visual inspection. Applications rapidly become more complex and often with more demanding real timeExpand
Cross-Domain Modeling and Optimization of High-Speed Visual Servo Systems
TLDR
This work applies an axiomatic design method to the design of high-speed visual servo systems, such that cross-domain couplings are explicitly modeled and subsequently eliminated when possible. Expand
Towards verified construction of correct and optimised GPU software
TLDR
The vision is presented of an integration of push-button formal verification techniques into a Model Driven Engineering workflow, and how steps in that direction are planned in the coming five years. Expand
Implementation, modeling, and exploration of precision visual servo systems
This chapter introduces visual servoing systems, and motives their applications in semi-structured environments. The challenges of designing precision visual servo systems are discussed. To overcomeExpand
Human Factors in Augmented Reality Environments
Advances in hardware and networking have made possible a wide use of augmented reality (AR) technologies. However, simply putting those hardware and technologies together does not make a good systemExpand
...
1
2
...

References

SHOWING 1-10 OF 139 REFERENCES
Skeletons and Asynchronous RPC for Embedded Data- and Task Parallel Image Processing
TLDR
This paper describes how to exploit task parallelism using an asynchronous remote procedure call (RPC) system, optimized for low-memory and sparsely connected systems such as smart cameras. Expand
SKiPPER: A Skeleton-Based Parallel Programming Environment for Real-Time Image Processing Applications
TLDR
This paper presents SKiPPER, a programming environment dedicated to the fast prototyping of parallel vision algorithms on MIMD-DM platforms based upon the concept of algorithmic skeletons, which can be executed on any sequential platform to check the correctness of the parallel algorithm. Expand
A methodology for the development and the support of massively parallel programs
TLDR
This work presents a methodology to easily write efficient, high performance and portable massively parallel programs, based on the definition of a new explicitly parallel programming language, namely P 3 L, and of a set of compiling tools that perform automatic adaptation of the program features to the target architecture hardware. Expand
Data- and Task Parallel Image Processing on a Mixed SIMD-ILP Platform using Skeletons and Asynchronous RPC
TLDR
This paper describes the asynchronous remote proce- dure call (RPC) system, optimized for low-memory and sparsely connected systems such as smart cameras, which uses a futures(13)-like model to present a normal imperative C-interface to the user in which the skeleton calls are im- plicitly parallelized and pipelined. Expand
Design and programming of embedded multiprocessors: an interface-centric approach
We present design technology for the structured design and programming of embedded multi-processor systems. It comprises a task-level interface that can be used both for developing parallelExpand
Algorithmic skeletons for stream programming in embedded heterogeneous parallel image processing applications
TLDR
This paper presents a C-like skeleton implementation language, PEPCI, that uses term rewriting and partial evaluation to specify skeletons for parallel C dialects, and provides a stream programming language that is better tailored to the user as well as the underlying architecture. Expand
APPLICATION DRIVEN DESIGN OF EMBEDDED REAL-TIME IMAGE PROCESSORS
TLDR
This paper will shortly review the archetypes of image processing architectures and their support, after which it will elaborate on a hard and software design framework for embedded image processors that is able to schedule the inherent data and task parallelism in an application in such a way, that a balance is found for both data andtask parallel parts of the application software. Expand
Algorithmic Skeletons: Structured Management of Parallel Computation
TLDR
This book introduces a new approach to the design and implementation of software systems which will help users of large scale parallel systems coordinate many concurrent activities toward a single goal and proposes a selection of independent algorithmic skeletons which describes the structure of a particular style of algorithm. Expand
A Skeleton Library
TLDR
The idea is to offer typical parallel programming patterns as polymorphic higher-order functions which are efficiently implemented in parallel as a library which can easily be used by e.g. C and C++ programmers. Expand
Efficient Implementation of Image Processing Algorithms on Linear Processor Arrays Using the Data Parallel Language IDC
TLDR
A data parallel language succinctly designed for a virtual LPA, and also a compiler for an existing LPA is described and a guideline for pamllel SIMD linear array algorithm developnlent using the language is provided. Expand
...
1
2
3
4
5
...