Smart scribbles for image matting

Image matting is an ill-posed problem that usually requires additional user input, such as trimaps or scribbles. Drawing a fine trimap requires a large amount of user effort, while using scribbles can hardly obtain satisfactory alpha mattes for non-professional users. Some recent deep learning-based...

Full description

Saved in:
Bibliographic Details
Main Authors: XIN, Yang, QIAO, Yu, CHEN, Shaozhe, HE, Shengfeng, YIN, Baocai, ZHANG, Qiang, WEI, Xiaopeng, LAU, Rynson W. H.
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2021
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/7881
https://ink.library.smu.edu.sg/context/sis_research/article/8884/viewcontent/SmartScribbles_av.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-8884
record_format dspace
spelling sg-smu-ink.sis_research-88842024-02-16T09:23:41Z Smart scribbles for image matting XIN, Yang QIAO, Yu CHEN, Shaozhe HE, Shengfeng YIN, Baocai ZHANG, Qiang WEI, Xiaopeng LAU, Rynson W. H. Image matting is an ill-posed problem that usually requires additional user input, such as trimaps or scribbles. Drawing a fine trimap requires a large amount of user effort, while using scribbles can hardly obtain satisfactory alpha mattes for non-professional users. Some recent deep learning-based matting networks rely on large-scale composite datasets for training to improve performance, resulting in the occasional appearance of obvious artifacts when processing natural images. In this article, we explore the intrinsic relationship between user input and alpha mattes and strike a balance between user effort and the quality of alpha mattes. In particular, we propose an interactive framework, referred to as smart scribbles, to guide users to draw few scribbles on the input images to produce high-quality alpha mattes. It first infers the most informative regions of an image for drawing scribbles to indicate different categories (foreground, background, or unknown) and then spreads these scribbles (i.e., the category labels) to the rest of the image via our well-designed two-phase propagation. Both neighboring low-level affinities and high-level semantic features are considered during the propagation process. Our method can be optimized without large-scale matting datasets and exhibits more universality in real situations. Extensive experiments demonstrate that smart scribbles can produce more accurate alpha mattes with reduced additional input, compared to the state-of-the-art matting methods. 2021-01-01T08:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/7881 info:doi/10.1145/3408323 https://ink.library.smu.edu.sg/context/sis_research/article/8884/viewcontent/SmartScribbles_av.pdf http://creativecommons.org/licenses/by/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Image matting alpha matte markov chain deep learning label propagation Graphics and Human Computer Interfaces
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Image matting
alpha matte
markov chain
deep learning
label propagation
Graphics and Human Computer Interfaces
spellingShingle Image matting
alpha matte
markov chain
deep learning
label propagation
Graphics and Human Computer Interfaces
XIN, Yang
QIAO, Yu
CHEN, Shaozhe
HE, Shengfeng
YIN, Baocai
ZHANG, Qiang
WEI, Xiaopeng
LAU, Rynson W. H.
Smart scribbles for image matting
description Image matting is an ill-posed problem that usually requires additional user input, such as trimaps or scribbles. Drawing a fine trimap requires a large amount of user effort, while using scribbles can hardly obtain satisfactory alpha mattes for non-professional users. Some recent deep learning-based matting networks rely on large-scale composite datasets for training to improve performance, resulting in the occasional appearance of obvious artifacts when processing natural images. In this article, we explore the intrinsic relationship between user input and alpha mattes and strike a balance between user effort and the quality of alpha mattes. In particular, we propose an interactive framework, referred to as smart scribbles, to guide users to draw few scribbles on the input images to produce high-quality alpha mattes. It first infers the most informative regions of an image for drawing scribbles to indicate different categories (foreground, background, or unknown) and then spreads these scribbles (i.e., the category labels) to the rest of the image via our well-designed two-phase propagation. Both neighboring low-level affinities and high-level semantic features are considered during the propagation process. Our method can be optimized without large-scale matting datasets and exhibits more universality in real situations. Extensive experiments demonstrate that smart scribbles can produce more accurate alpha mattes with reduced additional input, compared to the state-of-the-art matting methods.
format text
author XIN, Yang
QIAO, Yu
CHEN, Shaozhe
HE, Shengfeng
YIN, Baocai
ZHANG, Qiang
WEI, Xiaopeng
LAU, Rynson W. H.
author_facet XIN, Yang
QIAO, Yu
CHEN, Shaozhe
HE, Shengfeng
YIN, Baocai
ZHANG, Qiang
WEI, Xiaopeng
LAU, Rynson W. H.
author_sort XIN, Yang
title Smart scribbles for image matting
title_short Smart scribbles for image matting
title_full Smart scribbles for image matting
title_fullStr Smart scribbles for image matting
title_full_unstemmed Smart scribbles for image matting
title_sort smart scribbles for image matting
publisher Institutional Knowledge at Singapore Management University
publishDate 2021
url https://ink.library.smu.edu.sg/sis_research/7881
https://ink.library.smu.edu.sg/context/sis_research/article/8884/viewcontent/SmartScribbles_av.pdf
_version_ 1794549701990678528