Image foreground extraction and its application to neural style transferстатья
Информация о цитировании статьи получена из
Scopus
Статья опубликована в журнале из списка Web of Science и/или Scopus
Дата последнего поиска статьи во внешних источниках: 17 ноября 2021 г.
Авторы:
Kitov V. ,
Ponomareva L.
Журнал:
CEUR Workshop Proceedings
Том:
2830
Год издания:
2021
Издательство:
M. Jeusfeld c/o Redaktion Sun SITE, Informatik V, RWTH Aachen
Местоположение издательства:
Aachen, Germany
Первая страница:
191
Последняя страница:
200
Аннотация:
Foreground extraction plays important role in different computer vision applications: photo enhancement, image classification and understanding, style transfer improvement and others. New images dataset with annotation into foreground/background is proposed. Several recent neural segmentation models are trained on this dataset to extract foreground automatically and their performance is compared. The benefits of automatic foreground extraction are demonstrated on style transfer task - a popular technique for automatic rendering of photo (or content image) in the style defined by the style image, for example - the painting of a famous artist. © 2021 Copyright for this paper by its authors.
Добавил в систему:
Китов Виктор Владимирович