https://github.com/ribin-baby/u2net_pytorch

pytorch implimentation of u2net architecture
https://github.com/ribin-baby/u2net_pytorch

background-removal computer-vision deeplearning image-matting python pytorch u2net

Last synced: 7 months ago
JSON representation

pytorch implimentation of u2net architecture

• Host: GitHub
• URL: https://github.com/ribin-baby/u2net_pytorch
• Owner: Ribin-Baby
• License: unlicense
• Created: 2024-01-18T04:43:11.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-01-18T18:21:46.000Z (over 1 year ago)
• Last Synced: 2025-01-16T04:22:11.799Z (9 months ago)
• Topics: background-removal, computer-vision, deeplearning, image-matting, python, pytorch, u2net
• Language: Jupyter Notebook
• Homepage:
• Size: 80 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          

# U2 NET








Fig.1 - U2Net Architecture






* **U2-Net** is a two-level nested U-structure architecture. It uses a novel ReSidual U-block (RSU) module to extract multi-scale features without degrading resolution, allowing the network to go deeper and attain high resolution without significantly increasing memory and computation cost.

* used for for salient object detection, image segmentation, Image Matting, background removal and other image2image modeling tasks.








Fig.2 - UNet or RSU Block






* U-Net is a U-shaped encoder-decoder architecture with residual connections between each layers. It captures contextual information and intricate detail.

* These U-Net blocks in U2Net architecture are called ReSidual U-block or RSU.

*  Example: we have trained an Image Matting model on [P3M-10k](https://paperswithcode.com/dataset/p3m-10k) dataset, and the results are given below.








Fig.3 - Image Matting  with U2-Net example






![training progress](docs/training_progress.gif)

Fig.4 - Image Matting with U2-Net training progress after each steps