Skip to content

Different approaches to speed up disparity calculation via concurrency

Notifications You must be signed in to change notification settings

Tetr4/ParallelDisparityCalculation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Different approaches to speed up disparity calculation

Note: The code for the stereolab application is required to run these code fragments.

Elephant

Factors

  • Runtime Environment:
    • Just-in-time compilation (Java)
    • Ahead-of-time compilation (C)
    • On graphics processing unit (CUDA)
  • Concurrency Strategy
    • Single thread (no concurrency)
    • 1 thread per pixel
    • 1 thread per image-chunk

Results

Hardware

  • Intel Xeon E3-1230 v3 (3,30GHz, 4 Cores, 8 Threads)
  • GeForce GTX 770

Parameters

  • Image-Resolution: 681 × 681 (see image above)
  • Window: 15 × 15
  • Tau-Max: 40

Results

About

Different approaches to speed up disparity calculation via concurrency

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published