Image matting atoc
Download as PPTX, PDF0 likes162 views
Image matting is the process of separating the foreground and background of an image by assigning each pixel an alpha value between 0 and 1 indicating its transparency. Traditionally, matting uses a trimap to classify pixels as foreground, background, or uncertain. Early sampling-based methods calculated alpha values based on feature distances of closest foreground and background pixels. More recent approaches use deep learning, where the first deep learning matting method in 2016 took local and non-local information as input, and the 2017 Deep Image Matting method used an RGB image and trimap as input in a fully deep learning framework.
![Image Matting is
Image foreground background
Only deciding alpha! [0,1]](https://image.slidesharecdn.com/imagemattingatocr-190314094527/85/Image-matting-atoc-3-320.jpg)
![Image Matting is
Image foreground background
Only deciding alpha! [0,1]](https://image.slidesharecdn.com/imagemattingatocr-190314094527/85/Image-matting-atoc-4-320.jpg)
Image matting atoc
- 1. NotKind Introduction for Image Matting Segmentation은 아니라구
- 3. Image Matting is Image foreground background Only deciding alpha! [0,1]
- 4. Image Matting is Image foreground background Only deciding alpha! [0,1]
- 5. Matting is commonly with Trimap! For decide Foreground/Uncertain/Background
- 6. History of Matting A long time ago, …
- 7. Approaches – Sampling Based > Mishima Method 1993 – patent Calculating closest F/B Distance for decide alpha : globally * Feature - First method of matting - Look feature globally - Developed for Blue screen matting
- 8. Approaches – Sampling Based > Knockout software - not sure 1998…? Calculating closest F/B Distance for decide alpha : locally sampled * Feature - First method of matting with Trimap
- 9. Jumping matting history Closed Form Matting 2006 - Locally Matting KNN Matting 2012 - nonLocal Matting
- 10. Approaches – Deep Learning Age > DCNN Matting - 2016 Mix Local-Nonlocal information * Feature - First method of matting with Deep Learning Input is RGB + KNN Matting + Closed Form Matting
- 11. Approaches – Deep Learning Age > Deep Image Matting - 2017 * Feature Fully Deep Learning Refinement Stage is for alternative Image Guide Filter Input is RGB + Trimap
- 12. Want more? http://people.inf.ethz.ch/aksoyy/ Meet ‘Yağız Aksoy’ who starts from 2012 or https://www.ecse.rpi.edu/~rjradke/cvfxcourse. html kind lecture for matting