Skip to main content

A library and command to extract the smiliest image from a video

Project description

Smiler

This is a library and CLI tool to extract the "smiliest" of frame from a video of people.

It was developed as part of Choirless as part of IBM Call for code.

Installation

% pip install choirless_smiler

Usage

Simple usage:

% smiler video.mp4 snapshot.jpg

Output image of people singing

It will do a pre-scan to determine the 5% most changed frames from their previous frame in order to just consider them. If you know the threshold of change you want to use you can use that. e.g.

The first time smiler runs it will download facial landmark data and store it in ~/.smiler location of this data and cache directory can be specified as arguments

% smiler video.mp4 snapshot.jpg --threshold 480000

Help

% smiler -h
usage: smiler [-h] [--verbose] [--threshold THRESHOLD]
              [--landmarks-url LANDMARKS_URL] [--cache-dir CACHE_DIR]
              [--quantile QUANTILE]
              video_fn image_fn

Save thumbnail of smiliest frame in video

positional arguments:
  video_fn              filename for video to analyse
  image_fn              filename for output thumbnail

optional arguments:
  -h, --help            show this help message and exit
  --verbose             verbose mode
  --threshold THRESHOLD
                        threshold of difference overwhich we analyse an image
  --landmarks-url LANDMARKS_URL
                        url of facial landmark file
  --cache-dir CACHE_DIR
                        local cache to store the landmark file in
  --quantile QUANTILE   quantile of images to analyse

Verbose mode

If you specify the --verbose flag to the CLI then smiler will print out a progress bar as it goes along to show you have it's doing.

Use as a library

Smiler can be imported and used in a library. You are responsible for supplying paths to the facial landmark data and model, but help functions in the module can help.

from choirless_smiler.smiler import Smiler, load_landmarks

landmarks_path = load_landmarks(landmarks_url, cache_dir)

smiler = Smiler(landmarks_path, model_path)

fg = smiler.frame_generator(video_fn)
threshold = smiler.calc_threshold(fg, quantile)
fg = smiler.frame_generator(video_fn)
ffg = smiler.filter_frames(fg, threshold)

smile_score, image = smiler.find_smiliest_frame(ffg)

Re-training

There are some scripts in the scripts directory in the Github repo to generate new images and to aid manual classification and retraining of the model.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

choirless_smiler-1.2.0.tar.gz (300.9 kB view details)

Uploaded Source

File details

Details for the file choirless_smiler-1.2.0.tar.gz.

File metadata

  • Download URL: choirless_smiler-1.2.0.tar.gz
  • Upload date:
  • Size: 300.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.24.0 setuptools/41.2.0 requests-toolbelt/0.9.1 tqdm/4.47.0 CPython/3.7.7

File hashes

Hashes for choirless_smiler-1.2.0.tar.gz
Algorithm Hash digest
SHA256 ca240efd52f8333d102eb0e6e0e8e04f5b00ff3837eb7f1fe08df7ad8ae57606
MD5 c0d3e100a541fb32c0baee4b2abe1d36
BLAKE2b-256 3e4aee37a1fdabbd21ff2a4001424ef05af7bc07039b6e17227262b38fcd3134

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page