iFRAME (inhomogeneous Filters Random Field And Maximum Entropy)

Experiment 2: Learning Sparse FRAME by Gibbs Sampler on Coefficients and Generative Boosting

Code and dataset

Contents
Case 1 : Cat
Case 2 : Wolf
Case 3 : Pigeon
Case 4 : Zebra
Case 5 : Deer
Case 6 : Hummingbird
Case 7 : Cougar
Case 8 : Lion
Case 9 : Tiger
Case 10 : Flamingo
Case 11 : Bike
Case 12 : Leopard

more synthesis

Case 1: Cat

Training images:

synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=100; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 2: Wolf

Training images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=100; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 3: Pigeon

Traning images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=100; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 4: Zebra

Training images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=103; sizeTemplatey=123; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 5: Deer

Training images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=100; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 6: Hummingbird

Training images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=100; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 7: Cougar

Training images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=100; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 8: Lion

Training images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=100; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 9: Tiger

Training images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=100; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 10: Flamingo

Training images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=127; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 11: Bike

Training images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=100; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30

Case 12: Leopard

Training images

Synthesis by learning sparse-Frame model

(click here for movie of learning process)

Sketch template

parameters setting:

General Parameters: nOrient = 16; sizeTemplatex=70; sizeTemplatey=100; GaborScaleList=[ 1.4, 1, 0.7, 0.5]; DoGScaleList =[18.90, 13.36]; LocationShiftLimit=2; OrientShiftLimit=1; interval=5; #Wavelet=600; nIteration= (interval) x (#Wavelet);
Gibbs Sampler Parameters: lambdaLearningRate = 0.1/sqrt(sigsq); 6x6 chains; sigsq=10; threshold_corrBB=0; lower_bound_rand = 0.001; upper_bound_rand = 0.999; c_val_list=-30:2:30