Natural Image Matting via Guided Contextual Attention

Abstract

Over the last few years, deep learning based approaches have achieved outstanding improvements in natural image matting. Many of these methods can generate visually plausible alpha estimations, but typically yield blurry structures or textures in the semitransparent area. This is due to the local ambiguity of transparent objects. One possible solution is to leverage the far-surrounding information to estimate the local opacity. Traditional affinity-based methods often suffer from the high computational complexity, which are not suitable for high resolution alpha estimation. Inspired by affinity-based method and the successes of contextual attention in inpainting, we develop a novel end-to-end approach for natural image matting with a guided contextual attention module, which is specifically designed for image matting. Guided contextual attention module directly propagates high-level opacity information globally based on the learned low-level affinity. The proposed method can mimic information flow of affinity-based methods and utilize rich features learned by deep neural networks simultaneously. Experiment results on Composition-1k testing set and alphamatting.com benchmark dataset demonstrate that our method outperforms state-of-the-art approaches in natural image matting. Code and models are available at https://github.com/Yaoyi-Li/GCA-Matting.

Deep Residual Network for High-Resolution Background Matting

“…Recently, deep learning has shown impressive matting technique results (Chen et al, 2018;Shen et al, 2016). Shen et al (2016) propose a generation of the trimap from portrait image using a deep neural network and also propose a matting layer (Li et Lu., 2020) which uses the forward and backward propagation strategy. Chen et al (2018) introduced an automatic human matting algorithm without feeding trimaps.…”

“…Context-Aware Matting (Hou & Liu, 2019) introduces double decoders to estimate alpha and foreground map. GCA Matting (Li et Lu., 2020) uses deep learning to take advantage of the usergenerated trimap by employing a trimap-guided attention mechanism. Sengupta et al (2020) introduced Background Matting where a further background image is captured to serve as a big cue for predicting the alpha matte and therefore the foreground layer.…”

Automatic Trimap Generation by a Multimodal Neural Network

“…Many deep learning-based alpha matting methods have been proposed recently. Some of them require only photos as input [2,3], while others require photos and trimaps [1,4,5,6,7].…”

Improved Minimum Spanning Tree based Image Segmentation with Guided Matting

“…The matting algorithm can split the overall objects from a complex background in an image although it cannot detect the detailed information of the foreground, and make up the lack of graph-based algorithms in this regard [22][23]. The Bayesian matting algorithm was proposed by Chuang and Curless, they improved the Bayesian matting algorithm as most of pixels to be detected contain mixed colors which are the combination of foreground and background [24].…”