FaceSwap

A python implementation of an end-to-end pipeline for swapping faces from images and videos using classical and deep learning approach.

Run Instructions

python Wrapper.py --src $Path_for_source_image --dst $Path_for_destination_image --result Path_for storing_output_images/video --video $Path_for_input_video --method $METHOD_TYPE --mode $MODDE_TYPE

$MEHTOD_TYPE : 1 -> deltriangle; 2-> tps; 3 -> prnet

$MODE_TYPE : 1 -> Swap faces in two images ; 2 -> Swap face in video with an image; 3 -> Swap faces in a video

Classical Approach

Pipeline

Facial Landmarks detection

Inbuilt library dlib is used which is based on SVM face detector.

Face Warping using Triangulation and Thin Plate Spline

Deep Learning Approach

PRNet

This was implemented as part of CMSC733 and for detailed report refer here