2018 IEEE Visual Communications and Image Processing (VCIP) 2018
DOI: 10.1109/vcip.2018.8698689
|View full text |Cite
|
Sign up to set email alerts
|

Interactive Style Transfer: Towards Styling User-Specified Object

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
6
0

Year Published

2019
2019
2024
2024

Publication Types

Select...
6
2

Relationship

0
8

Authors

Journals

citations
Cited by 11 publications
(7 citation statements)
references
References 4 publications
0
6
0
Order By: Relevance
“…e CNN model needs a huge amount of data sets that are very difficult in medical based image processing. For this problem, the image generation is used to increase the computer vision [24]. A new engine is developed to exploit the network to confine the synthetic information.…”
Section: Related Workmentioning
confidence: 99%
“…e CNN model needs a huge amount of data sets that are very difficult in medical based image processing. For this problem, the image generation is used to increase the computer vision [24]. A new engine is developed to exploit the network to confine the synthetic information.…”
Section: Related Workmentioning
confidence: 99%
“…• Bo Ren is the corresponding author (rb@nankai.edu.cn). [45]. Although these NST methods, even with global adjustment, can create artworks, they can not provide fine-grained controls in artistic editing, and are not in line with natural drawing logic.…”
Section: Introductionmentioning
confidence: 99%
“…Procedural content generation (PCG) is the systematic automation of producing content merging user-generated sprites, audio, and visuals using algorithmic approaches in order to create enhanced, diverse and automatic content in a fast and transitive way [4]. Neural style transfer is one of the PCG approaches that create diverse artistic styles [5][6][7][8][9][10][11] and get remarkable results, mainly in the image processing domain. Neural style transfer has become a highly researched topic in recent years due to deep learning practices [5,11].…”
Section: Introductionmentioning
confidence: 99%
“…One solution to this problem is to associate semantic labeling of the input and style images with maximizing the subregion mapping, as proposed in [6]. In [7], a different approach in which the foreground segmentation was combined with the neural style transfer was used. The main idea was to style only the user-specified object, which was done by first styling the whole image and then separately segmenting out the object to overlap the segmented and styled image further.…”
Section: Introductionmentioning
confidence: 99%