A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Filters
A Generative Adversarial Framework for Optimizing Image Matting and Harmonization Simultaneously
[article]
2021
arXiv
pre-print
Preserved Fulltext
We propose a new Generative Adversarial (GAN) framework which optimizing the matting network and the harmonization network based on a self-attention discriminator. ...
Image matting and image harmonization are two important tasks in image composition. ...
Therefore the image matting may be more accurate and suitable for the new background, and the harmonization will also benefit from the more accurate input alpha matte. ...
arXiv:2108.06087v1
fatcat:lljbmwgjojajfe7nte7lpoqvq4
Foreground color prediction through inverse compositing
2021
2021 IEEE Winter Conference on Applications of Computer Vision (WACV)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Our method outperforms the state-of-the-art in color estimation for natural image matting and show that the recurrent nature of our method allows users to easily change candidate solutions that lead to ...
In recent years, advances in deep learning have led to many natural image matting algorithms that have achieved outstanding performance in a fully automatic manner. ...
[20] and use the predicted alpha and foreground colors to composite a new image with a plain background. ...
doi:10.1109/wacv48630.2021.00165
fatcat:ftdjx3l7dvfwnotmtuvdbhbaqi
AlphaGAN: Generative adversarial networks for natural image matting
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We present the first generative adversarial network (GAN) for natural image matting. ...
Our novel generator network is trained to predict visually appealing alphas with the addition of the adversarial loss from the discriminator that is trained to classify well-composited images. ...
We would also like to thank the authors of the Deep Image Matting [33] paper for providing us with their training dataset and their evaluation code. ...
arXiv:1807.10088v1
fatcat:xeb7v5u2cjh7fhjaisd6t2aloa
Instance Segmentation based Semantic Matting for Compositing Applications
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Image compositing is a key step in film making and image editing that aims to segment a foreground object and combine it with a new background. ...
that can be used for image editing task. ...
Natural Image Matting Image matting estimates an alpha matte for an interesting object by generating an opacity value for each pixel in the image. ...
arXiv:1904.05457v1
fatcat:vtx2iixpzrgwdn2rhzyr3zq6cu
Deep Image Matting
[article]
2017
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Image matting is a fundamental computer vision problem and has many applications. ...
In addition, we also create a large-scale image matting dataset including 49300 training images and 1000 testing images. ...
To train our matting network, we create a larger dataset by compositing objects from real images onto new backgrounds. ...
arXiv:1703.03872v3
fatcat:55rboswwofb23eku7ydyahi3iu
Deep Image Matting
2017
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Image matting is a fundamental computer vision problem and has many applications. ...
In addition, we also create a large-scale image matting dataset including 49300 training images and 1000 testing images. ...
To train our matting network, we create a larger dataset by compositing objects from real images onto new backgrounds. ...
doi:10.1109/cvpr.2017.41
dblp:conf/cvpr/XuPCH17
fatcat:tgjtpqvkkvhvdh5go7scsbrlpm
Deep Automatic Portrait Matting
[chapter]
2016
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2016; you can also visit the original URL.
The file type is application/pdf.
Our method considers not only image semantic prediction but also pixel-level image matte optimization. A new portrait image dataset is constructed with our labeled matting ground truth. ...
We propose an automatic image matting method for portrait images. This method does not need user interaction, which was however essential in most previous approaches. ...
This work is supported by a grant from the Research Grants Council of the Hong Kong SAR (project No. 2150760) and by the National Science Foundation China, under Grant 61133009. ...
doi:10.1007/978-3-319-46448-0_6
fatcat:3qa524r6xba7zoseqxxytydq3i
Improving alpha matte with depth information
2009
IEICE Electronics Express
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Our matting method improves existing natural alpha matting methods by adding depth information. ...
In this paper, we propose an automatic alpha matting method that improves the quality of alpha matte using a depth camera. ...
This matting enables soft extraction of a foreground from a given image so that it allows a seamless composition of a foreground image with a new background image. ...
doi:10.1587/elex.6.1602
fatcat:wdfwevva6rfy7gazrc6hvsnpvi
Automatic Natural Video Matting with Depth
2007
15th Pacific Conference on Computer Graphics and Applications (PG'07)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2008; you can also visit the original URL.
The file type is application/pdf.
In addition, we show that we can improve the results from natural matting algorithms by adding a depth channel. ...
We look at existing singleframe matting techniques and present a method that improves upon them by adding depth information acquired by a time-of-flight range scanner. ...
The matte is used to separate the foreground from the image in order to put a new background behind it. ...
doi:10.1109/pg.2007.52
dblp:conf/pg/WangFYDY07
fatcat:gcamnucy5ves3i7pioxzggkz6q
Disentangled Image Matting
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Most previous image matting methods require a roughly-specificed trimap as input, and estimate fractional alpha values for all pixels that are in the unknown region of the trimap. ...
We propose AdaMatting, a new end-to-end matting framework that disentangles this problem into two sub-tasks: trimap adaptation and alpha estimation. ...
Most of the existing image matting methods consider image matting as a single regression task, which ignores the classification nature resided in image matting. ...
arXiv:1909.04686v1
fatcat:lri2hzyp45d5jd35ffhey36xqi
Background Matting: The World Is Your Green Screen
2020
2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Figure 1 : Using a handheld smartphone camera, we capture two images of a scene, one with the subject and one without. ...
We employ a deep network with an adversarial loss to recover alpha matte and foreground color. We composite the result onto a novel background. ...
The resulting mattes for real images can still have significant artifacts, particularly evident when compositing onto a new background. ...
doi:10.1109/cvpr42600.2020.00236
dblp:conf/cvpr/SenguptaJCSK20
fatcat:srixk6dunzhz7hxbz5cisjs55i
Multi-View Image Matting and Compositing Using Trimap Sharing for Natural 3-D Scene Generation
2008
2008 3DTV Conference: The True Vision - Capture, Transmission and Display of 3D Video
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
In this paper, we propose a new estimation scheme for multi-view alpha mattes by sharing the trimap of the reference view. ...
Recently, a digital matting algorithm has been developed to create an alpha matte using a well-focused image generated from multi-view images. ...
In this paper, we propose a new digital matting algorithm to estimate multi-view alpha mattes using a multi-view camera system. ...
doi:10.1109/3dtv.2008.4547892
fatcat:tzkb77zzxffdpo6mroubll7mqa
Matting Algorithm with Improved Portrait Details for Images with Complex Backgrounds
2024
Applied Sciences
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2024; you can also visit the original URL.
The file type is application/pdf.
In this paper, a new matting algorithm with improved portrait details for images with complex backgrounds (MORLIPO) is proposed. ...
and provide a more accurate prediction of the alpha matte of the current frame for the video image. ...
Acknowledgments: The authors would like to express their gratitude to the editors and anonymous reviewers for their comments and suggestions. ...
doi:10.3390/app14051942
fatcat:b4cmec2vkzgtfanwag5nkejucm
Human Instance Matting via Mutual Guidance and Multi-Instance Refinement
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
This paper introduces a new matting task called human instance matting (HIM), which requires the pertinent model to automatically predict a precise alpha matte for each human instance. ...
Finally, we construct a HIM benchmark for evaluation, which comprises of both synthetic and natural benchmark images. ...
Related Work
Matting Natural Image Matting. Image matting is a pixel-level task, aiming to extract alpha matte for a foreground object. ...
arXiv:2205.10767v1
fatcat:nbnwf2p2ojfkflzposo7j23akq
Learning-Based Sampling for Natural Image Matting
2019
2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
The goal of natural image matting is the estimation of opacities of a user-defined foreground object that is essential in creating realistic composite imagery. ...
Natural matting is a challenging process due to the high number of unknowns in the mathematical modeling of the problem, namely the opacities as well as the foreground and background layer colors, while ...
Introduction Natural image matting is the estimation of the accurate soft transitions between a user-defined foreground and the background of an image. ...
doi:10.1109/cvpr.2019.00317
dblp:conf/cvpr/TangAOGA19
fatcat:gkawnpdg25aqjn5u5lghgtrpnm
« Previous
Showing results 1 — 15 out of 11,607 results