Home

Awesome

Hybrid Images

Code style: black

Hybrid images demonstrate how certain frequencies dominate at certain distances. Most notably, high frequencies dominate when the object is close you you, while lower frequencies dominate when it is further away.

We can demonstrate the effect of hybrid images by applying a Gaussian blur to an image in order to produce a low-pass filtered equivalent.

cat cat hpf

For a different (but aligned) image we again Gaussian blur, but then subtract the low-pass image from the original in order to produce a high pass image.

cat cat hpf

These two images can then be summed together to produce a hybrid image as shown below. The image is downsampled to give the effect of moving the object closer/further away from your view point.

Hybrid effect

Installation

(Optional) Setup a virtual environment to install the necessary packages.

virtualenv .venv
source .venv/bin/activate

Install the packages listed in the requirements file.

pip install -r requirements.txt

Run the program!

Usage

This program has a command-line interface built up using click. An auto-generated help message is printed with the -h/--help flag.

> python hybrid.py -h
Usage: hybrid.py [OPTIONS] COMMAND [ARGS]...

  Hybrid image demonstration program.

Options:
  -h, --help  Show this message and exit.

Commands:
  hybrid  Create hybrid image from two source images.
  kernel  Demonstrate the effect of kernel size.
  sobel   Perform sobel edge detection.

> python hybrid.py hybrid -h
Usage: hybrid.py hybrid [OPTIONS] IMAGES...

  Create hybrid image from two source images.

Options:
  -o, --output TEXT        Output file.
  -c, --cutoff INTEGER...  High/low cutoff frequencies.
  -v, --visual             Generate visualisation.
  -f, --fourier            Use fourier convolution.
  -h, --help               Show this message and exit.

> python hybrid.py kernel -h
Usage: hybrid.py kernel [OPTIONS] IMAGE

  Demonstrate the effect of kernel size.

Options:
  -o, --output TEXT      Output file.
  -s, --size INTEGER...  Kernel dimensions.
  -h, --help             Show this message and exit.

> python hybrid.py sobel -h
Usage: hybrid.py sobel [OPTIONS] IMAGE

  Perform sobel edge detection.

Options:
  -o, --output TEXT  Output file.
  -h, --help         Show this message and exit.