cmaruan/lfcodec

Light_Field_Codec

Clone and build

To build this repository, you should have cmake 3.15 installed in your system.

First, clone the repo and its modules dependencies.

$ mkdir repo
$ git clone https://github.com/IDSRosler/Light-Field-Codec.git repo
$ cd repo
$ git submodule init
$ git submodule update

Then, make a directory to build the project binaries

$ mkdir build
$ cd build
$ cmake ..
$ make 

From the build directory, run the binary ./lfcodec.

Running tests

To execute the tests, run the binary build/tst/lfcodec-test.

Encoder Parameters (CLI)

Use: -flag value

Example: ./bin.exe -input /temp/

Directories

• input string: *.ppm dir to be encoded
• output string: output dir for the encoded *.ppm

Base block dimensions

• blx int: block size in x dimension
• bly int: block size in y dimension
• blu int: block size in u dimension
• blv int: block size in v dimension

Quantization Parameters

• qp float: linear quantization parameter for the quantization 4d-volum
• qx int: quantization weight in x dimension
• qy int: quantization weight in y dimension
• qu int: quantization weight in u dimension
• qv int: quantization weight in v dimension

Light Field dimensions

• lfx int: light field size in x dimension
• lfy int: light field size in y dimension
• lfu int: light field size in u dimension
• lfv int: light field size in v dimension

Scripts

• Scripts to run the Light Field Codec for different QPs and different dataset

Run simulation.py:

1. In the simulation.py, change the input_file variable and the output_file variable
   • input_file: Directory with .ppm files to be encoded
   • output_file: Output directory for the encoded *.ppm
2. Execute
   • taskset -c 0 python3 simulation.py