Awesome
<div align="center"> <img src="doc/image/logo.gif" width="300"/> </div> <div align="center">- A Photography Portrait Matting Benchmark - </div> <p align="center"> <a href="#news">News</a> | <a href="#introduction">Introduction</a> | <a href="#download">Download</a> | <a href="#license">License</a> | <a href="#citation">Citation</a> | <a href="#contact">Contact</a> </p> <img src="doc/image/title.jpg" width="100%">News
- [Jul 20 2021] PPM-100 Benchmark is Released!
The benchmark with 100 finly-annotated, high-resolution images (PPM-100) is released.
Introduction
PPM is a portrait matting benchmark with the following characteristics:
- Fine Annotation - All images are labeled and checked carefully.
- Natural Background - All images use the original background without replacement.
- Rich Diversity - The images cover full/half body and various postures.
- High Resolution - The resolution of images is between 1080p and 4k.
Below is an example image:
<img src="doc/image/example.png" width="100%">Download
Currently, PPM-100 used in the MODNet paper is available.
Note that few images used in the MODNet paper are replaced by similar images due to license issues.
You can download PPM-100 from:
Google Drive | 百度网盘 (提取码: PPMB)
License
All original portrait images in PPM are from Flickr and constrained by Flickr Creative Commons License (Commercial use & mods allowed).
All annotated alpha mattes in PPM are released under the Creative Commons Attribution NonCommercial ShareAlike 4.0 license.
Citation
If you use this PPM benckmark, please cite:
@InProceedings{MODNet,
author = {Zhanghan Ke and Jiayu Sun and Kaican Li and Qiong Yan and Rynson W.H. Lau},
title = {MODNet: Real-Time Trimap-Free Portrait Matting via Objective Decomposition},
booktitle = {AAAI},
year = {2022},
}
Contact
This repository is currently maintained by Zhanghan Ke (@ZHKKKe).
If there is any question, please contact kezhanghan@outlook.com.