ViTAE-Transformer/P3M-Net
The official repo for [IJCV'23] "Rethinking Portrait Matting with Privacy Preserving"
This helps graphic designers and content creators cleanly separate people from their backgrounds in photos, even when faces are blurred for privacy. You input an image with a person, and it outputs a precise 'cut-out' of the person (an alpha matte) that can be easily placed onto new backgrounds. This is ideal for anyone who regularly needs to remove photo backgrounds while respecting privacy.
112 stars. No commits in the last 6 months.
Use this if you need to extract human subjects from images, especially when faces might be intentionally blurred or anonymized, and you require high-quality, fine-detail cut-outs.
Not ideal if your main task is to remove backgrounds from objects or scenes without people, or if privacy-preserving training is not a concern for your specific image matting needs.
Stars
112
Forks
9
Language
Python
License
MIT
Category
Last pushed
Mar 30, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/ViTAE-Transformer/P3M-Net"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
patrikhuber/eos
A lightweight 3D Morphable Face Model library in modern C++
Uason-Chen/CTR-GCN
[ICCV2021] Official code for "Channel-wise Topology Refinement Graph Convolution for...
PeterL1n/BackgroundMattingV2
Real-Time High-Resolution Background Matting
PeterL1n/RobustVideoMatting
Robust Video Matting in PyTorch, TensorFlow, TensorFlow.js, ONNX, CoreML!
Zielon/MICA
MICA - Towards Metrical Reconstruction of Human Faces [ECCV2022]