Statistical analysis of sample values to approximate the final pixel value

Jérôme BUISINE 950feeab7d Merge branch 'release/v0.1.1'		4 anni fa
generate	95cba96b9b Use of modules dependency	4 anni fa
modules @ d5de038bdc	95cba96b9b Use of modules dependency	4 anni fa
notebooks	87b8d18661 Write Keras result script	5 anni fa
others	95cba96b9b Use of modules dependency	4 anni fa
reconstruct	95cba96b9b Use of modules dependency	4 anni fa
.gitignore	ab86ef6206 Documentation and ignore files updates	4 anni fa
.gitmodules	95cba96b9b Use of modules dependency	4 anni fa
LICENSE	d0ebc63822 Update of documentation	4 anni fa
README.md	ab86ef6206 Documentation and ignore files updates	4 anni fa
__init__.py	beb4716c41 Add of feature choices	4 anni fa
compare_images.py	95cba96b9b Use of modules dependency	4 anni fa
custom_config.py	adf8908549 Update of keras config	4 anni fa
features.py	95cba96b9b Use of modules dependency	4 anni fa
generate_data.sh	cccee3b412 Update of Keras scripts	5 anni fa
run.sh	8590bd14ac Update of run script	4 anni fa
run_keras.sh	a6aa09a009 Keras run script update	4 anni fa
train_model.py	018d8e801f Update of train model script	4 anni fa
train_model_keras.py	95cba96b9b Use of modules dependency	4 anni fa

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

README.md