Faces

😔 → 😀

A Python wrapper around FaceApp .

Installation

$ pip install git+https://github.com/vasilysinitsin/Faces.git or manually clone this repo.

Basic Usage

With file,

import faces

sad = open('sad.jpg', 'rb')
image = faces.FaceAppImage(file=sad)
happy = image.apply_filter('smile', cropped=True)

with URL

import faces

old_rockfeller = 'https://upload.wikimedia.org/wikipedia/commons/6/6f/John_D._Rockefeller_1885.jpg'
image = faces.FaceAppImage(url=old_rockfeller)
young_rockfeller = image.apply_filter('young')

or with FaceApp code and device id

import faces

code_of_me_uploaded = '20170517181457gflf'
my_device_id = '12345678'
image = faces.FaceAppImage(code=code_of_me_uploaded, device_id=my_device_id)
brad_pitt = image.apply_filter('hot')

Handling Exceptions

try:
    image = faces.FaceAppImage(...)
except faces.ImageHasNoFaces:
    print('Your face is not recognized. Are you an alian?')
except faces.BadImageType:
    print('This image is not valid. Get some good bytes.')
except faces.BaseFacesException:
    print('Some unknown wrong things happened.')

try:
    result = image.apply_filter('young, rich and powerful')
except faces.BadFilterID:
    print('Too cool filter to exist.')

Known filters

are stored in faces.KNOWN_FILTERS list. API is not documented so expect them to change or brake over time. You can track which of them works on Travis CI latest build info. It is scheduled to be updated daily.

Advanced features

Dumping and rebuilding class from json

# It is handy when you have uploader-worker application and have to pass data between.

image = faces.FaceAppImage(...)
json_string = image.to_json() # type(json_string) == str

"""
...pass to different machine
"""

rebuilt_image = faces.FaceAppImage.from_json(json_string) # type(rebuilt_image) == faces.FaceAppImage

Yes, it's that easy. Now create something cool!

Tests and API status

This module hardly relies on undocumented FaceApp API. Tests are scheduled to run daily on Travis CI to ensure API works as expected.