https://github.com/hlfshell/facial_recognition
An educational project to learn about facial keypoint recognition with a CNN architecture
https://github.com/hlfshell/facial_recognition
Last synced: about 1 year ago
JSON representation
An educational project to learn about facial keypoint recognition with a CNN architecture
- Host: GitHub
- URL: https://github.com/hlfshell/facial_recognition
- Owner: hlfshell
- Created: 2020-01-28T04:01:11.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2022-11-22T05:21:30.000Z (over 3 years ago)
- Last Synced: 2025-02-09T17:14:54.866Z (over 1 year ago)
- Language: Python
- Size: 39.1 KB
- Stars: 1
- Watchers: 2
- Forks: 0
- Open Issues: 10
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Facial Keypoint Project
This project follows a project offered by Udacity's Computer Vision Nanodegree program, as described via their [Github repos](https://github.com/udacity/P1_Facial_Keypoints). While that project uses a series of guided notebooks, which I did reference, I created this project as a standalone exploration of pytorch as well.
As described, the goal of the project is to utilize a dataset of facial keypoints and build a model that
1. Given an image, find each face within the image, and
2. Given a set a face found within that image, identify each facial keypoint
# Files
The following key files are within this project:
* `journal.md` - This is an open ended set of notes about the process, hardships, and quesitons raised during the course of this project. This is a mapping of the journey throughout.
* `model.py` - Where the pytorch model is defined.
* `training.py` - This file handles the training process for the model.
* `dataset.py` - This file acts as a tool to download and prepare the dataset for training.
* `inference.py` - This loads a trained model and passes it through our model, outputting the results.