Dynamic control and prototyping of parallel algorithms for intermediate- and high-level vision

Wallace, Andrew M.; Michaelson, Greg; McAndrew, Patrick; Waugh, Kevin; Austin, W. J.

doi:10.1109/2.121474

Cited by 14 publications

(13 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have achieved respectable parallel performance for hand translations of functional vision algorithms [9,11] and have developed a general purpose parallelising compiler to automate the translation process [3] which has shown promising results with simple programs.…”

Section: Discussionmentioning

confidence: 99%

“…We give brief examples of each algorithm in SML using the higher order function constructs that we have indicated can be implemented in parallel. Although we do not explicitly state, for each example, how the parallel implementation is derived from the prototype, we have substantial experience in this process [9,11] using 0ccam2 as the target language based on a Meiko transputer surface. We have shown that if we can write a functional language program using our standard set of higher order functions then we can develop an equivalent parallel implementation based on the parallel constructs for these functions.…”

Section: Fun Filter P [] = [] Imentioning

confidence: 99%

See 1 more Smart Citation

Prototyping Parallel Algorithms using Standard ML.

Scaife¹,

Michaelson²,

Wallace³

1995

Procedings of the British Machine Vision Conference 1995

View full text Add to dashboard Cite

We have been developing techniques for deriving parallel implementations of vision algorithms from prototypes written in a functional language (SML). Initially, we analysed simple, well understood algorithms to allow the prototyping methodology to be investigated in a predictable environment. Subsequently, we have extended our approach to more difficult cases such as edge tracking which present problems for parallel system development. Here we demonstrate the power and generality of our approach to parallel algorithm development by a representative set of vision algorithms encoded in SML.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Fun Filter P [] = [] Imentioning

confidence: 99%

Prototyping Parallel Algorithms using Standard ML.

Scaife¹,

Michaelson²,

Wallace³

1995

Procedings of the British Machine Vision Conference 1995

View full text Add to dashboard Cite

show abstract

“…Following the work of Illingworth and Kittler [18] and Wallace [19], a different kind of algorithm was developed to improve performance over those which have been described above. Such an algorithm relies on an iterative technique of refinement based on dynamic parameter range reduction.…”

Section: Algorithm 4: Iterative Refinementmentioning

confidence: 99%

“…The algorithm devised for this experiment is similar to that by Wallace et al [19] and is based on an iterative technique whereby most of the processing is done on areas of the accumulator which are likely to possess the highest peaks. The aim of the technique is to lower the amount of work done by starting off the computation at a very coarse accumulator quantization and then gradually focusing in on a few selected parameter ranges at each iteration [11,18].…”

Section: Algorithm 4: Iterative Refinementmentioning

confidence: 99%

Performance of the Hough transform on a distributed memory multiprocessor

Underhill

Atiquzzaman

Ophel

1999

Microprocessors and Microsystems

View full text Add to dashboard Cite

The Hough transform is a projection-based transform which can be used to detect shapes in images. One of the disadvantages of the transform is its requirement for large amounts of computing power. Parallel machines have given programmers the potential for incredible computing power. To obtain maximum performance from parallel machines, parallel algorithms should be designed to reflect the architecture of the parallel machine. The work reported in this paper compares the performance obtained in running several parallel versions of the Hough transform on a Fujitsu AP1000 distributed memory multiprocessor. ᭧ 1999 Published by Elsevier Science B.V.

show abstract

“…Parallel algorithms for 3D model-based matching from perspective images exist already [11]. The same direct solution can be used to solve 3-point and 4-point perspective problems allowing a number of difference scene primitives to be handled, for example trihedral junctions and line-arc combinations by 3-point perspective inversion and line-line combinations by 4-point perspective inversion.…”

Section: Further Workmentioning

confidence: 99%

Recognition and Location by Parallel Pose Clustering.

Austin¹,

Wallace²

1995

Procedings of the British Machine Vision Conference 1995

View full text Add to dashboard Cite

This paper describes the recognition and location of 3D object models from depth-based data using a parallel pose clustering algorithm. We describe a leader-based partitional clustering algorithm and demonstrate a successful parallel implementation of this. Results are presented for a variety of synthetic and real scene data. We also consider how the basic approach can be extended to recognise objects from a single 2D intensity image by perspective inversion. Our eventual aim is to combine such a dual data approach within a single parallel system.

show abstract

Dynamic control and prototyping of parallel algorithms for intermediate- and high-level vision

Cited by 14 publications

References 8 publications

Prototyping Parallel Algorithms using Standard ML.

Prototyping Parallel Algorithms using Standard ML.

Performance of the Hough transform on a distributed memory multiprocessor

Recognition and Location by Parallel Pose Clustering.

Contact Info

Product

Resources

About