A software environment for parallel computer vision

Jamieson, Leah H.; Delp, Edward J.; Wang, C.-C.; Li, J.; Weil, Frank

doi:10.1109/2.121478

Cited by 32 publications

(10 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The single characteristic that makes Beowulf clusters favorable, however, is that they deliver better price-performance with respect to alternative systems. From these properties, in combination with the fact that many references exist that show significant performance gains for a multitude of different image processing applications (see, e.g., [18][19][20][21]), we conclude that Beowulf clusters constitute the most appropriate target platforms for our specific needs.…”

Section: Hardware Architecturesmentioning

confidence: 97%

User transparency: a fully sequential programming model for efficient data parallel image processing

Seinstra

Koelma

2004

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYAlthough many image processing applications are ideally suited for parallel implementation, most researchers in imaging do not benefit from high-performance computing on a daily basis. Essentially, this is due to the fact that no parallelization tools exist that truly match the image processing researcher's frame of reference. As it is unrealistic to expect imaging researchers to become experts in parallel computing, tools must be provided to allow them to develop high-performance applications in a highly familiar manner. In an attempt to provide such a tool, we have designed a software architecture that allows transparent (i.e. sequential) implementation of data parallel imaging applications for execution on homogeneous distributed memory MIMD-style multicomputers. This paper presents an extensive overview of the design rationale behind the software architecture, and gives an assessment of the architecture's effectiveness in providing significant performance gains. In particular, we describe the implementation and automatic parallelization of three well-known example applications that contain many fundamental imaging operations: (1) template matching; (2) multi-baseline stereo vision; and (3) line detection. Based on experimental results we conclude that our software architecture constitutes a powerful and user-friendly tool for obtaining high performance in many important image processing research areas.

show abstract

Section: Hardware Architecturesmentioning

confidence: 97%

User transparency: a fully sequential programming model for efficient data parallel image processing

Seinstra

Koelma

2004

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…The authors opted for a static code analysis approach. Earlier work in [8] again built upon the idea of patterns in communication and computation as programming abstractions for algorithms. The paper discusses the problem of task scheduling, for which we use an NP-complete optimal algorithm, which they solve with heuristic-guided graph isomorphism algorithms in order to reuse subgraph architectural mappings that are known to work well.…”

Section: Related Workmentioning

confidence: 99%

High-performance SIMT code generation in an active visual effects library

Cornwall

Howes

Parsonage

et al. 2009

Proceedings of the 6th ACM Conference on Computing Frontiers

View full text Add to dashboard Cite

SIMT (Single-Instruction Multiple-Thread) is an emerging programming paradigm for high-performance computational accelerators, pioneered in current and next generation GPUs and hybrid CPUs. We present a domain-specific activelibrary supported approach to SIMT code generation and optimisation in the field of visual effects. Our approach uses high-level metadata and runtime context to guide and to ensure the correctness of optimisation-driven code transformations and to implement runtime-context-sensitive optimisations. Our advanced optimisations require no analysis of the original C++ kernel code and deliver 1.3x to 6.6x speedups over syntax-directed translation on GeForce 8800 GTX and GTX 260 GPUs with two commercial visual effects.

show abstract

“…In this work, a rather 'brute force' object-recognition technique is presented that yields reliable detection results while offering the potential to be implemented on fast, parallelprocessing graphics hardware [8,11,13]. Given the 3-D geometry of an object, the proposed recognition scheme autonomously determines the object's apparent size, image coordinates, and its pose in an uncalibrated image.…”

Section: Introductionmentioning

confidence: 99%

Geometry-based automatic object localization and 3-D pose detection

Magnor

Proceedings Fifth IEEE Southwest Symposium on Image Analysis and Interpretation

View full text Add to dashboard Cite

show abstract

A software environment for parallel computer vision

Cited by 32 publications

References 2 publications

User transparency: a fully sequential programming model for efficient data parallel image processing

User transparency: a fully sequential programming model for efficient data parallel image processing

High-performance SIMT code generation in an active visual effects library

Geometry-based automatic object localization and 3-D pose detection

Contact Info

Product

Resources

About