The use of new technologies inspired us to work for this project and also the learning we were supposed to get from this project has inspired us a lot to do this.

What it does

We present an image-based face swapping algorithm, which can be used to replace the face in the reference image with the same facial shape and features as the input face. It can be used for identification of deep fake faces for security purpose

How I built it

First we have face alignment  which is based on a group of detected facial points so that the input face and the reference face are consistent in size and posture. Next an image warping algorithm is used which is based on triangulation and it is presented to adjust the reference face and the background according to the aligned input faces. In order to achieve more accurate face swapping effect, a face parsing algorithm is used to get the accurate detection of the face-ROIs, and then the face-ROI in the reference image is replaced with the input face-ROI. Finally, a Poisson image editing algorithm is used to check the boundary processing and color correction between the replacement region and the original background, and then the final face swapping result is obtained.

Challenges I ran into

The major challenge which I faced working on this project was learning about the computer vision algorithms and managing the time to give the perfect output.

Accomplishments that I'm proud of

I went through many research paper of the same project but the fact and novelty in my project is the prediction rate of my model is much more higher and accurate then the all model which has been discovered till date.

What I learned

I learned about how to manage time and learn new skills in a short period of time. I also learned about some of the computer vision algorithms like image warping algorithm , face parsing algorithm , Poisson image editing algorithm and I also learned about some approaches to solve a computer vision problem like replacement-based approach, model-based approach and learning based approach. Overall it was a nice learning experience and working experience.

What's next for Realistic Face swapping Using computer Vision

If we talk about the bussiness aspect of Face swapping then it can be used by different government and non government organisation to detect the face with High accuracy for safety purpose and we can use face swapping in entertainment industry as well to make high level 8D graphic movies. Thank you Its all from my side at the end I enjoyed and learned a lot from this hacakthon

Built With

Share this project: