realify
Turn quick face sketches into photorealistic 2D portraits and interactive 3D models
Science Score: 44.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Committers with academic emails
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (6.1%) to scientific vocabulary
Keywords
Repository
Turn quick face sketches into photorealistic 2D portraits and interactive 3D models
Basic Info
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Topics
Metadata Files
README.md
REALIFY
Transform your roughest scribbles to realistic face potraits, and turn those into an interactive 3d model.
Demo Video
https://github.com/user-attachments/assets/23916501-df8a-43fc-8484-706a3fc0679b
Realify UI
Realify 2D pipeline
Realify 3D model
Other Examples
How It Works
- Scribble: Draw freely on the 2D canvas. Includes draw, fill, and erase tools.
- realify: Use the “realify” button to transform scribble into realistic photograph of a face
- REALIFY: Click “REALIFY” to convert your photograph into a 3D model
- Iterate: Draw another face and repeat!
Quick Start
Prerequisites
- Conda venv
- Libraries in requirement.txt
- Controlnet diffusion model
Program Setup
bash
python3 -m pip install -r requirements.txt
source ~/miniconda3/etc/profile.d/conda.sh
conda activate face3d
Frontend Setup
bash
cd frontend
npm start
Backend Setup
bash
cd backend
python3 -m uvicorn main:app --reload
Architecture
Frontend
- React: Clean, responsive UI
- Three.js: Interactive 3D model
- Figma: Initial UI design
- Html Css, Node.js: Styling and animations
Backend
- FastAPI: Quick API frameowrk
- Mediapipe: 3D inferencing
- Bash: Convenient installation
Owner
- Name: Michael Chang
- Login: Michael660-max
- Kind: user
- Repositories: 3
- Profile: https://github.com/Michael660-max
uni student who likes coding.
Citation (CITATION.bib)
@misc{zhang2023adding,
title = {Adding Conditional Control to Text-to-Image Diffusion Models},
author = {Lvmin Zhang and Anyi Rao and Maneesh Agrawala},
booktitle = {IEEE International Conference on Computer Vision (ICCV)},
year = {2023},
}
GitHub Events
Total
- Push event: 7
Last Year
- Push event: 7
Committers
Last synced: 7 months ago
Top Committers
| Name | Commits | |
|---|---|---|
| Michael Chang | M****4@g****m | 33 |
Issues and Pull Requests
Last synced: 7 months ago
Dependencies
- 1316 dependencies
- @react-three/drei ^10.0.8
- @react-three/fiber ^9.1.2
- @testing-library/dom ^10.4.0
- @testing-library/jest-dom ^6.6.3
- @testing-library/react ^16.3.0
- @testing-library/user-event ^13.5.0
- react ^19.1.0
- react-dom ^19.1.0
- react-scripts 5.0.1
- three ^0.176.0
- web-vitals ^2.1.4
- @types/draco3d 1.4.10
- @types/offscreencanvas 2019.7.3
- @types/webxr 0.5.22
- draco3d 1.5.7
- fflate 0.6.10
- potpack 1.0.2
- three 0.176.0
- three-stdlib 2.36.0
- three ^0.176.0
- three-stdlib ^2.36.0
- accelerate >=0.21.0
- controlnet-aux *
- diffusers >=0.24.0
- fastapi *
- huggingface-hub >=0.21.0
- numpy *
- open3d *
- opencv-python *
- pillow *
- python-multipart *
- safetensors *
- timm *
- torch *
- torchvision *
- transformers >=4.51.0
- uvicorn *