Real-Time-Video-Processing

Clone the repository and install the requirement by:
pip install -r requirements.txt

Face_Recognition: Pre-Encoded Face Detection and Tracking

Change directory to Face_Recognition and depending on os run:
For windows:
python Face_Recognition.py

For linux:
python3 Face_Recognition.py

Adding new face from images and also from live video. Recognizing faces from images and live video. If face is not pre-recorded then unknown is shown.
face_recognition_dataset: dataset for creating face recognition encodings
haarcascades: haarcascades file
recognize_faces: directory to put images in which faces are to be recognized
Add_New_Face.py: Functions to add new faces to existing encodings
Face_Recognition.py: Menu driven programm to add new face from photos folder(the name of folder should be name of person and the folder should be located in face_recognition_dataset), add new face from live vide, recognize face from images(images should be in recognize_faces folder), and recognize face from webcam video.

Object_Detection

Change directory to Object_Detection and depending on os run:
For windows:
python Object_Detection.py

For linux:
python3 Object_Detection.py

Object detection using pre-trained caffe model with classes: "background", "aeroplane", "bicycle", "bird", "boat", "bottle", "bus", "car", "cat", "chair", "cow", "diningtable", "dog", "horse", "motorbike", "person", "pottedplant", "sheep", "sofa", "train", "tvmonitor"
Object_Detection.py: Detecting objects from webcam video
caffe_models: pre-trained models used

Face_Liveness: Detecting wheteher a real face is there a a spoofed face(image of face)

Run the Live_Model_Demo.ipynb in jupyter

caffe_models: pre-trained models for face_detection
live_dataset: real and fake face images
training_videos: Either use pre-recorded real and fake face videos, or from live video
Live_Model_Demo.ipynb: Final demo for detecting real and fake faces
Liveness_Data_Generation.py: Extracting faces from videos
Liveness_Model.py: Model Architiecture for training
Model_Train.ipynb: Training the model

Combined_Model_Demo.ipynb

Combined object detection, face recogniiton and face liveness detection in live video stream.

Refrences

https://www.pyimagesearch.com/start-here/