https://github.com/gdjkhp/ama-facial-recognition

Free to Use AI Face Recognition (O_O)
https://github.com/gdjkhp/ama-facial-recognition

cctv face-detection face-recognition flask imgbeddings mediapipe openai-clip opencv opencv-python pyqt5 python surveillance

Last synced: 9 months ago
JSON representation

Free to Use AI Face Recognition (O_O)

• Host: GitHub
• URL: https://github.com/gdjkhp/ama-facial-recognition
• Owner: GDjkhp
• License: wtfpl
• Created: 2024-12-10T05:00:59.000Z (over 1 year ago)
• Default Branch: master
• Last Pushed: 2025-01-20T02:40:49.000Z (over 1 year ago)
• Last Synced: 2025-01-20T03:27:59.492Z (over 1 year ago)
• Topics: cctv, face-detection, face-recognition, flask, imgbeddings, mediapipe, openai-clip, opencv, opencv-python, pyqt5, python, surveillance
• Language: Python
• Homepage: https://gdjkhp.github.io/ama-facial-recognition/
• Size: 33.2 KB
• Stars: 1
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Funding: .github/FUNDING.yml
  • License: LICENSE

Awesome Lists containing this project

README

          


AMA Facial Recognition: Digital Records-logs Management System

A Thesis Presented to the Faculty of AMA Computer College Lucena Campus, Lucena City




In Partial Fulfilment of the Requirements for the Degree Bachelor of Science in Computer Science







John Emmanuel S. Marco




John Kennedy H. Peña






Copyright 2024, The Karakters Kompany



Developers' Notes

fuck i need money so bad i need to publish this messed up paper tas need pa ma-bookbind kasi requirements potek nayan i hate college so much im dedicating this project sa lahat ng mga nasa kolehiyo pa kasi grabe dinanas ko (>_<)






if you enjoyed the time you wasted then it's not a waste of time <3




donate kayo if you find this repo helpful, minimum lang sweldo ko sa trabaho :3




halos lahat ng python script ay AI generated ni OpenAI at Claude, kaya kung may tanong kayo, wag sakin, tanong nyo sa kanila XD




## para saan to?

para mas mapadali mag monitor ng mga tao sa paligid kasi may trust issues si eman (spy ng china)

### mga use case:

* pwedeng gawing biometrics at cctv surveillance

* nakakatamad magsulat sa logbook, merong pa bang contact-tracing?

* para mawalan ng trabaho si manong guard >:D

## pano to gamitin?

### madali lang to guys need mo lang neto:

* computer na may internet connection (preferably [windows](https://www.microsoft.com/en-us/windows) pero pwede rin linux or mac)

* [python](https://www.python.org) na hindi kalumaan (need mo rin [vs code](https://code.visualstudio.com) para mas maganda mag edit)

* analog or [ip cameras](https://en.wikipedia.org/wiki/IP_camera) (mas malinaw mas maganda)

### kung ready na lahat, sundan mo to:

1. install **ALL** required libraries: [requirements.txt](https://gdjkhp.github.io/ama-facial-recognition/requirements.txt) or pwede rin mano-mano good luck (hint: pip install)

2. download mo **LAHAT** ng mga script: [download as zip](https://github.com/GDjkhp/ama-facial-recognition/archive/refs/heads/master.zip)

3. enroll mo mukha mo: set-up mo si server ([server_capture.py](https://gdjkhp.github.io/ama-facial-recognition/server_capture.py)) then picture picture ([mp_client.py](https://gdjkhp.github.io/ama-facial-recognition/mp_client.py))

4. train mo si AI: extract mo lahat ng mga mukha ([readdb.py](https://gdjkhp.github.io/ama-facial-recognition/readdb.py)) then train ([mp_train.py](https://gdjkhp.github.io/ama-facial-recognition/mp_train.py))

5. finally try mo na sya congrats: ([mp_deploy.py](https://gdjkhp.github.io/ama-facial-recognition/mp_deploy.py))

## recipe

* `imgbeddings` vector embeddings tokenizer for images to assign names for each faces, powered by openai's clip model (credit goes to this [video](https://www.youtube.com/watch?v=Y0dLgtF4IHM))

* `mediapipe` 3D face detection by google (credit goes to eman)

* `opencv-contrib-python` cameras, yeah just for the cameras

* `pyqt5` gui nothing too fancy

* `numpy` numbers and array things

* `pillow` image processing

* `flask`, `requests` server/client network stuff (optional)

## format ng mga files at structure ng source code

* `/*.py` lahat ng logic nandito

* `/requirements.txt` lahat ng required python libraries nandito

* `/json/*.json` base64 image data ng mga mukha + other information na nakalap ni server

* `/received_images/{name}/*.png or *.jpg` lahat ng mga mukha na ni train ni AI

* `/face_embeddings.json` AI weight pagkatapos i-train

* `/face_detection_logs.csv` dito nagtatala ng date and time at camera forda person

* `/LICENSE` DO WHAT THE FUCK YOU WANT TO PUBLIC LICENSE

* `/README.md` itong binabasa mo

hindi rito kasama node.js server at client code, gamitin nyo nalang server_capture.py at mp_client.py pero kung need nyo to, chat nyo nalang si eman (O_o)