KNN Matting

@article{Chen2013KNNM,
  title={KNN Matting},
  author={Qifeng Chen and Dingzeyu Li and Chi-Keung Tang},
  journal={2012 IEEE Conference on Computer Vision and Pattern Recognition},
  year={2013},
  pages={869-876}
}
We are interested in a general alpha matting approach for the simultaneous extraction of multiple image layers; each layer may have disjoint segments for material matting not limited to foreground mattes typical of natural image matting. [] Key Method Our matting technique, aptly called KNN matting, capitalizes on the nonlocal principle by usingK nearest neighbors (KNN) in matching nonlocal neighborhoods, and contributes a simple and fast algorithm giving competitive results with sparse user markups.
Color clustering matting
TLDR
The local color line model which is based on the assumption of linear color clustering within a small local window, to nonlocal feature space neighborhood is extended and new affinity matrix is defined to achieve better clustering.
Natural image matting via adaptive local and nonlocal sample clustering
TLDR
This work develops an adaptive sample clustering criterion to automatically assign either local or nonlocal neighborhood to each pixel, and improves the nonlocal clustering performance by introducing a new feature selection parameter to choose preferred feature space for different images in a fully automatic way.
KNN-based color line model for image matting
TLDR
This paper proposes a K-nearest neighbors (KNN)-based color line model that combines and preserves the advantages of both the above assumptions and indicates that the experimental matting results are of comparable or higher quality than those obtained by the existing methods based on the above two assumptions.
Image Matting for Sparse User Input by Iterative Refinement
TLDR
This work proposes an iterative extension to the class of affinity matting techniques that focuses on achieving satisfactory results with extremely sparse input, e.g. a handful of individual pixel labels.
A comprehensive study on digital image matting
TLDR
A unified framework for digital image matting is introduced, which provides the possibility of obtaining a better understanding and direction of further improvement for image matts problem and proves the feasibility of the proposed framework.
Alpha matting using robust color sampling and fully connected conditional random fields
TLDR
A novel approach to extract foreground elements from an image through color and opacity estimations, which considers available samples in a search window of variable size for each unknown pixel.
Semantic Image Matting
TLDR
This work considers and learns 20 classes of matting patterns, and proposes to extend the conventional trimap to semantic trimap, and adds a multi-class discriminator to regularize the alpha prediction at semantic level, and content-sensitive weights to balance different regularization losses.
Unified Interactive Image Matting
TLDR
This work proposes UIM, a Unified Interactive image Matting method, which solves the limitations and achieves satisfying matting results for any scenario, and leverages multiple types of user interaction to avoid the ambiguity of multiple matting targets.
Image Alpha Matting via Residual Convolutional Grid Network
TLDR
A residual convolutional grid network for alpha matting is proposed, which deals with those matting problems well and has a performance comparable to the best image matting method in the literature.
Matting with Sequential Pair Selection Using Graph Transduction
TLDR
This research proposes a new sampling approach that brings relevant samples to every pixel with an unknown alpha value; this yields accurate alpha maps and presents two new formulations for objective functions used to assess the suitability of the chosen samples.
...
...