# iFRAME (inhomogeneous Filters Random Field And Maximum Entropy)

## Experiment 2: Wavelet Selection by Generative Boosting

data

Learning sequences by generative boosting are displayed. The first row shows the evolution of the sketch templates. Each selected Gabor wavelet is illustrated by a bar, and each DoG by a circle. The template is the superposition of Gabor wavelets of 4 scales and DoG wavelets of 2 scales, where smaller scales appear darker. The second row shows the synthesized images that are generated when the number of selected wavelets n = 1, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, and 700.

## Case 1: Hummingbird

## Case 2: Wolf

## Case 3: Bike

## Case 4: Flamingo

## Case 5: Falcon

**Typical parameters: **

image size is 100x100. The number of basis functions is 700. The allowed range of perturbation in location is 2 pixels. The allowed range of perturbation in rotation is pi/16.
GaborScaleList=[ 1.4, 1,0.7,0.5]. DoGScaleList =[18.90,13.36]. interval=3. 6x6 chains.

threshold_corrBB = 0; lower_bound_rand = 0.001; upper_bound_rand = 0.999;c_val_list=-25:2:25.