https://github.com/parkyunsu/fashion-u-want-virtual-try-on
Fashion U Want : A High Resolution Virtual Try-On system leveraging deep learning to seamlessly visualize clothes on human models.
https://github.com/parkyunsu/fashion-u-want-virtual-try-on
deep-learning deeplabv3plus detectron2 graphonomy hr-viton openpose python u2net
Last synced: 3 months ago
JSON representation
Fashion U Want : A High Resolution Virtual Try-On system leveraging deep learning to seamlessly visualize clothes on human models.
- Host: GitHub
- URL: https://github.com/parkyunsu/fashion-u-want-virtual-try-on
- Owner: PARKYUNSU
- Created: 2024-11-13T12:51:37.000Z (7 months ago)
- Default Branch: main
- Last Pushed: 2025-03-03T16:17:24.000Z (3 months ago)
- Last Synced: 2025-03-03T17:26:49.772Z (3 months ago)
- Topics: deep-learning, deeplabv3plus, detectron2, graphonomy, hr-viton, openpose, python, u2net
- Language: Python
- Homepage:
- Size: 194 MB
- Stars: 1
- Watchers: 1
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Fashion-U-Want: High Resolution Virtual Try-On
PyTorch implementation of Fashion U Want: High Resolution Virtual Try-On with GAN models
### Tech Stack
---
Fashion-U-Want is a high-resolution virtual try-on system that uses deep learning to overlay clothing onto a person's image. This framework processes input images through ``Try_TRYON.ipynb``, extracting clothing masks, pose information, and segmentation data to generate realistic try-on results—all without requiring manual installation of dependencies.
## Project Duration
**2024.11.13 - 2024.11.20**
Yunsu Park
Minju Lee
Myoungjin Son
## Presentation
The presentation deck is available in the `deck` folder: [Fashion_U_Want_presentation.pdf](https://github.com/PARKYUNSU/Fashion-U-Want-Virtual-Try-On/blob/main/deck/Fashion_U_Want_presentiaon.pdf).
## How to Use
1. Click the "Open in Colab" button below.
2. Follow the instructions in the notebook to upload input images and generate try-on result.
3. View the output directly within the notebook.
## Fashion-U-Want: Demo
Click the badge below to run the demo:
[](https://colab.research.google.com/github/PARKYUNSU/Fashion-U-Want-Virtual-Try-On/blob/main/Try_TRYON.ipynb)
## Results
## References
### Pose Estimation
- **OpenPose**: Realtime multi-person 2D pose estimation using part affinity fields.
Repository: [https://github.com/CMU-Perceptual-Computing-Lab/openpose](https://github.com/CMU-Perceptual-Computing-Lab/openpose)
### Semantic Segmentation
- **U2Net**: U-Net based architecture for salient object detection and segmentation.
Repository: [https://github.com/xuebinqin/U-2-Net](https://github.com/xuebinqin/U-2-Net)
- **EfficientNet B7**: High-performance image segmentation model, pre-trained on ImageNet.
Repository: [https://github.com/Karel911/TRACER](https://github.com/Karel911/TRACER)
### Dense Pose
- **Detectron2**: Object detection and human parsing framework for research and production.
Repository: [https://github.com/facebookresearch/detectron2](https://github.com/facebookresearch/detectron2)
### Image Parsing
- **DeepLabV3+**: Encoder-decoder with atrous separable convolution for semantic image segmentation.
Repository: [https://github.com/tensorflow/models/tree/master/research/deeplab](https://github.com/tensorflow/models/tree/master/research/deeplab)
### Human Parsing
- **Graphonomy**: Universal human parsing framework for garment understanding.
Repository: [https://github.com/Gaoyiminggithub/Graphonomy](https://github.com/Gaoyiminggithub/Graphonomy)
### Virtual Try-On
- **HR-VITON**: High-resolution virtual try-on for clothing.
Repository: [https://github.com/sangyun884/HR-VITON](https://github.com/sangyun884/HR-VITON)