This app generates the average image from a given set of images.
The images received as inputs are converted to n-dimensional vectors.
The converted to vectors set of images gets its barycenter computed and written in the static/output_images folder.
- nodejs and npm
Run npm install to install this app's dependencies
Run npm run compile to compile the app from TypeScript to JavaScript ES5.
Run npm start to run the app.
An example set is provided in the app static folder, holding picture with theme city.
Users can simply run npm start to regenerate the average image of this demo set.
Users can also empty the static/input_images folder and fill it in with new pictures.
Update the value override to false if images with same name in output_images folder should have their name set to unique name.
Update the outWidth and outHeight values with integers values to change the output pictures' dimensions.
Update the outExt value to change the extension of generated images (supported jpg, jpeg and png).
Then the script can be run again to generate new average images from new sets of images.
Here is a low resolution image generated from a set of 10 city pictures :
