Skip to content

Exposure fusion: Course project for the How to Write Fast Numerical Code lecture at ETH Zurich

Notifications You must be signed in to change notification settings

jmiserez/fastcode14

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

fastcode14

Fast Exposure Fusion implementation

  • Conversion of images to uncompressed TIFF: $ convert -compress none file.jpg $ convert -resize 10% -compress none file.jpg

Directories

/src: /01ref_matlab: 1:1 matlab implementation /consolidated: Different versions of the source code for benchmarking /driver: Benchmarking support code /perf: PMU support code, cost model support code /standalone: completely standalone version of the code, with separate Makefile /unused: unused and unsupported code /reference: contains the original reference Matlab code /testdata: contains sample input and output images for verification /utils: contains dynamorio code to measure ILP

Running benchmarks

fusion_driver.py configures, builds and runs different versions of code for benchmarking.

cd src/ ./fusion_driver.py

Options can be changed in the file fusion_driver.py, all Makefiles and binaries are generated automatically.

The optimized versions of the source code are in the src/consolidated/ folder, which are then combined using the appropriate Make.config files.

Standalone version with debug outputs

/src/standalone contains a standalone version of the program, it can be run with:

cd src/standalone/ make ./run_example.sh

This will run the program once and store all intermediate images from the algorithm.

About

Exposure fusion: Course project for the How to Write Fast Numerical Code lecture at ETH Zurich

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published