Create a system that automatically manipulates images found online.
I wanted to play with images and text.
So I downloaded a bunch of alpaca images (because they’re really funny) from Shutterstock and using one of Darius Kazemi’s corpora dictionaries (mood adjectives) manipulated them with the text.
Scrape images from Shutterstock
Add text from the dictionary using Python Pillow library
Create a gif
After I had all the images ready, I played around with ImageMagick library for Python and created few gifs using the below command:
convert -delay 35 -loop 0 *.png alpaca.gif
35 is a gif speed, if the number is lower, the gif speed will be faster, *.png are all the images that should be taken (I had 102) and alpaca.gif is the name of the gif that I’m creating.
Github repository can be found here.
Note: I had a hard time scraping images from Flickr and Google. Since they would work better (no watermarks), I’m still going to update the code once I make it work.