benjamindyer/face-morphing

👯‍♀️ We are more alike than different - morphing one face to another

Face Morphing

Create a morphing sequences betwen two faces.

Input: Two images containing faces
Output: A video showing the fluid transformation from one face to the other

Requirements

numpy
scikit_image
opencv_python
Pillow
skimage
dlib

Getting Started

Test with demo images

A photo of Jennie from Blackpink	A photo of Rihanna

Generate a morphing animation video sequence

python3 code/__init__.py --img1 images/aligned_images/jennie.png --img2 images/aligned_images/rih.png --output output.mp4

Test with your own images

1. Put your images in Images folder

2. Auto align faces with python code/utils/align_images.py images/ images/aligned_images --output_size=1024 This will look for faces in the images - crop out, align (center the nose and make the eyes horizontal), and then rescale the resulting images and save them in "aligned_images" folder.

3. Run code/__init__.py above on your aligned face images with arg --img1 and --img2.

Key Features

1. Detect and auto align faces in images (Optional for face morphing)
2. Generate corresponding features points between the two images using Dlib's Facial Landmark Detection
3. Calculate the triangular mesh with Delaunay Triangulation for each intermediate shape
4. Warp the two input images towards the intermediate shape, perform cross-dissolve and obtain intermediate images each frame

More Results

To Do

Morph multiple images into a complete sequence
Morph with body landmarks

Citations

Adivces on working with facial landmarks with dlib and opencv https://www.pyimagesearch.com/2017/04/03/facial-landmarks-dlib-opencv-python/