Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

0.2.1 (2020-10-30)

Bug Fixes

• Fixing the release task and some inconsistencies that were causing the download_data() function to raise a ValueError.

0.2.0 (2020-10-30)

Miscellaneous

• Due to PyPi’s upload limit of ~100MB, we cannot bundle pre-trained models along with the built package. We are now building a process for acquiring these models around an included function that will attempt to fetch the latest released models from GitHub releases.

  The method download_data() should be a quick initial setup task when attempting to use this module. This task will download the necessary data to use the included detectors and encoders.

  More details about what is necessary for this release process is and should (in the future) be documented in the facelift._data module.

0.1.0 (2020-10-27)

Miscellaneous

• The initial release doesn’t have a super detailed list of introduced features or bugfixes as this project was pulled together from other side projects I’ve had in the past. Below I’ll list the important features that we are starting out with. Future additions should result in a history of news fragments that get aggregated into this changelog.

  Starting features:

  1. Face feature detection with a few bundled models.

     • Basic face feature detection (eyes and nose)

     • Partial face feature detection (trained model produced by dlib)

     • Full face feature detection (third party trained model)

  2. Face recognition with a bundled ResNet produced by dlib to produce face encoding.

     • Includes basic Euclidean distance scoring to find similar faces.

  3. Wrappers for OpenCV frame capturing.

     • Generators for frames from written media files.

     • Generators for frames from streaming devices (webcams).

  4. Wrappers for OpenCV windows.

     • Context managers for named window management.

  5. Wrappers for OpenCV common frame transformations

     • Scaling, resizing, rotating, cutting, copying, etc…

  6. Wrappers for OpenCV canvas drawing features

     • Helper functions for drawing points, lines, polygons, text, etc…

  7. Example helpers module for basic face normalization.

     • Gives a basic re-implementation of dlib’s get_face_chip() method.