Statistical analysis of sample values to approximate the final pixel value

Jérôme BUISINE 2a671c77e4 Merge tag 'v0.1.1' into develop 5 lat temu
generate 95cba96b9b Use of modules dependency 5 lat temu
modules @ d5de038bdc 95cba96b9b Use of modules dependency 5 lat temu
notebooks 87b8d18661 Write Keras result script 5 lat temu
others 95cba96b9b Use of modules dependency 5 lat temu
reconstruct 95cba96b9b Use of modules dependency 5 lat temu
.gitignore ab86ef6206 Documentation and ignore files updates 5 lat temu
.gitmodules 95cba96b9b Use of modules dependency 5 lat temu
LICENSE d0ebc63822 Update of documentation 5 lat temu
README.md ab86ef6206 Documentation and ignore files updates 5 lat temu
__init__.py beb4716c41 Add of feature choices 5 lat temu
compare_images.py 95cba96b9b Use of modules dependency 5 lat temu
custom_config.py adf8908549 Update of keras config 5 lat temu
features.py 95cba96b9b Use of modules dependency 5 lat temu
generate_data.sh cccee3b412 Update of Keras scripts 5 lat temu
run.sh 8590bd14ac Update of run script 5 lat temu
run_keras.sh a6aa09a009 Keras run script update 5 lat temu
train_model.py 018d8e801f Update of train model script 5 lat temu
train_model_keras.py 95cba96b9b Use of modules dependency 5 lat temu

README.md

Sample Analysis

Description

The aim of this project is to predict the mean pixel value from monte carlo process rendering in synthesis images using only few samples information in input for model.

Data

Data are all scenes samples information obtained during the rendering process.

For each pixel we have a list of all grey value estimated (samples).

Models

List of models tested :

  • Ridge Regression
  • SGD
  • SVR (with rbf kernel)

How to use

First you need to contact jerome.buisine@univ-littoral.fr in order to get datatset version. The dataset is not available with this source code.

python make_dataset.py --n 10 --each_row 8 --each_column 8
python reconstruct.py --scene Scene1 --model_path saved_models/Model1.joblib --n 10 --image_name output.png

License

The MIT license