quereste/deepfake-for-the-good
Official repository of paper "Deepfake for the Good: Generating Avatars through Face-Swapping with Implicit Deepfake Generation"
This tool helps creative professionals or digital artists generate realistic 3D or 4D (video) avatars by combining a single 2D photo of a person with a video of a base face model. You provide a reference photo and a base avatar video, and it outputs a deepfake video or 3D model where the base avatar takes on the appearance of the person in your photo, even allowing for expression changes. It's ideal for visual content creators, animators, or marketing specialists.
No commits in the last 6 months.
Use this if you need to create realistic 3D or 4D digital avatars from a single reference image for virtual assistants, digital doubles, or interactive experiences.
Not ideal if you require a simple, one-click solution without needing to integrate multiple external tools or have limited computational resources (especially for 4D models).
Stars
36
Forks
2
Language
Jupyter Notebook
License
—
Category
Last pushed
Aug 04, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/quereste/deepfake-for-the-good"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
OPTML-Group/Unlearn-Saliency
[ICLR24 (Spotlight)] "SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in...
Shilin-LU/VINE
[ICLR 2025] "Robust Watermarking Using Generative Priors Against Image Editing: From...
WindVChen/DiffAttack
An unrestricted attack based on diffusion models that can achieve both good transferability and...
koninik/DiffusionPen
Official PyTorch Implementation of "DiffusionPen: Towards Controlling the Style of Handwritten...
Wuyxin/DISC
(ICML 2023) Discover and Cure: Concept-aware Mitigation of Spurious Correlation