Remote Data Mining And Management Job In Data Science And Analytics

Deep Fake / Face Swap Video Processing development project

Find more Data Mining And Management remote jobs posted recently Worldwide

Were looking for a developer that has experience with producing deep fakes for doing a face swap on a Linux server preferably on AWS with a GPU Cuda instance.
Functionality: User uploads a face photo or video of themselves, and their face appears on the face of the person in our base video that is on the server. The base video can be a static video that is preprocessed so it has its learning model prepared. The users uploaded face needs to go onto the face in the video.
It can either be a faceswap, or a deep fake combination. The mouth should move with the preprocessed video.

DeepFake/FaceSwap POC requirements:

Were looking to have a server solution that replaces a face in a pre recorded video with an uploaded photo or video of a user.

Functionality:
-We have a prerecorded mp4 video on the server that is a person that is more or less head on in the shot. (eg. driving a car with a head on shot of them). This video could be something different for this test just as a proof of concept though, in case you have a sample video that works. This prerecorded video can be 10-15seconds long.

-The prerecorded video needs to have a process identified in creating the deep learning needed for using it as the base video. So identifying how to process that as the base video, and the estimate for time required to process it

-The user uploads a photo of themselves (or video of 5 seconds) and the users face then is processed and replaced on to the face of the person in the pre recorded video

-Were looking for an api call for this to happen so we can pass in the user asset (photo) to the backend script that does the face replacement of the video

-The end result video is then returned through the api call ideally

-Also looking for a time estimate for how long this process takes

-Optional adjustments - how much face blending, or lighting adjustments, anything else that goes into making it seamless into the prerecorded video.

-Should run on an AWS instance / ami with GPU, preferably Linux
About the recuiter
Member since Mar 14, 2020
Roopa Selvanaya
from Vojvodina, Yugoslavia

Skills & Expertise Required

Computer Vision 

Candidate shortlisted and hiredHiring open till - Sep 25, 2020

Work from Anywhere

40 hrs / week

Hourly Type

Remote Job

$19.47

Cost

Looking for help? Checkout our video tutorial
How to search and apply for jobs

How to apply? Do you have more questions about the Job?
See frequently asked questions

Similar Projects

Raspberry Pi • AWS Computer Vision Facial Recognition App

Looking for expert-level Firmware Development specialist to code Raspberry Pi and AWS Computer Vision facial recognition with sensors and camera.

Robust Image Matching Algorithm

I am looking for a freelancer who has very deep understanding of Computer Vision and Machine Learning who can develop a robust image matching algorithm. For the project, I have tried feature based/keypoint matching algorithm including ORB, SIFT, Akaz...read more

Train Neural Network and build a CNN

Train Neural Network and build a CNN(Fastest: SSD mobileNet or similar ) model. Model should support on browser(JavaScript file can access this trained model for prediction). Should have some knowledge on WEBGL on browser level. (Its AR/VR with AI o...read more

Need algorithm to run YOLO on Raspberry Pi at 1 FPS

Understand and implement YOLO(You Look Only Once) v3 algorithm on Raspberry Pi 3B+
Optimize algorithm to achieve a decent FPS close to real-time(1fps)
Achieve accuracy of at least 90 % in object detection mostly to detect people from an overh...read more

Looking for someone who knows python - creating an emotion image from emotion recondition video

Hello,
Im looking for someone who can take a video which analyze emotions and create a still image from the analysis.
I need the still image to be an average emotion (many stills from opacity)