dongwu92/AutoPortraitMatting
A TensorFlow implementation of Fully Convolutional Networks for automatic portrait segmentation and matting.
Not currently ranked — collecting fresh signals.
star history
This repository provides a TensorFlow implementation of automatic portrait matting based on the paper ‘Deep Automatic Portrait Matting’. It uses a Fully Convolutional Network architecture to segment portraits from images and extract precise foreground boundaries with alpha mattes. The model was originally published at ECCV 2016 by Xiaoyong Shen et al.
Frequently asked
- What is dongwu92/AutoPortraitMatting?
- A TensorFlow implementation of Fully Convolutional Networks for automatic portrait segmentation and matting.
- Is AutoPortraitMatting open source?
- Yes — dongwu92/AutoPortraitMatting is open source, released under the Apache-2.0 license.
- What language is AutoPortraitMatting written in?
- dongwu92/AutoPortraitMatting is primarily written in Python.
- How popular is AutoPortraitMatting?
- dongwu92/AutoPortraitMatting has 477 stars on GitHub.
- Where can I find AutoPortraitMatting?
- dongwu92/AutoPortraitMatting is on GitHub at https://github.com/dongwu92/AutoPortraitMatting.