Map-inspired Art Image
碩士 === 國立成功大學 === 資訊工程學系 === 102 === This paper presents a map-inspired art rendering technique that embeds real figures into city maps while preserving their shapes and detailed features as well as road network topologies in city maps. Most previous researches make an effort to embed hidden figure...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2014
|
Online Access: | http://ndltd.ncl.edu.tw/handle/43436322843263566341 |
id |
ndltd-TW-102NCKU5392024 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-102NCKU53920242016-03-07T04:10:58Z http://ndltd.ncl.edu.tw/handle/43436322843263566341 Map-inspired Art Image 地圖風格的藝術影像 Che-HaoLi 李哲皓 碩士 國立成功大學 資訊工程學系 102 This paper presents a map-inspired art rendering technique that embeds real figures into city maps while preserving their shapes and detailed features as well as road network topologies in city maps. Most previous researches make an effort to embed hidden figures into another image and these methods are usually utilized for natural scenes but not for full structure pictures such as city maps. Therefore, the embedded results in the city maps potentially create unnatural virtual road networks or texture artifacts in both detailed features and shapes of city maps by ignoring the density control of map simplification or complication. In this paper, we propose a map-inspired reducing-and-increasing scheme to ameliorate these repulsive artifacts as well as enhance the shapes and detailed contents. The basic idea is to combine both reducing and increasing techniques, where the reducing approach removes non-significant roads and the increasing enhances local features to preserve the shape of salient objects, detailed features, and road network topologies. We show many map-inspired art images generated by our technique. We have tested the quality of the images in an extensive user study, showing better results than those by other state-of-the-art approaches. Tong-Yee Lee 李同益 2014 學位論文 ; thesis 39 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立成功大學 === 資訊工程學系 === 102 === This paper presents a map-inspired art rendering technique that embeds real figures into city maps while preserving their shapes and detailed features as well as road network topologies in city maps.
Most previous researches make an effort to embed hidden figures into another image and these methods are usually utilized for natural scenes but not for full structure pictures such as city maps. Therefore, the embedded results in the city maps potentially create unnatural virtual road networks or texture artifacts in both detailed features and shapes of city maps by ignoring the density control of map simplification or complication.
In this paper, we propose a map-inspired reducing-and-increasing scheme to ameliorate these repulsive artifacts as well as enhance the shapes and detailed contents. The basic idea is to combine both reducing and increasing techniques, where the reducing approach removes non-significant roads and the increasing enhances local features to preserve the shape of salient objects, detailed features, and road network topologies.
We show many map-inspired art images generated by our technique. We have tested the quality of the images in an extensive user study, showing better results than those by other state-of-the-art approaches.
|
author2 |
Tong-Yee Lee |
author_facet |
Tong-Yee Lee Che-HaoLi 李哲皓 |
author |
Che-HaoLi 李哲皓 |
spellingShingle |
Che-HaoLi 李哲皓 Map-inspired Art Image |
author_sort |
Che-HaoLi |
title |
Map-inspired Art Image |
title_short |
Map-inspired Art Image |
title_full |
Map-inspired Art Image |
title_fullStr |
Map-inspired Art Image |
title_full_unstemmed |
Map-inspired Art Image |
title_sort |
map-inspired art image |
publishDate |
2014 |
url |
http://ndltd.ncl.edu.tw/handle/43436322843263566341 |
work_keys_str_mv |
AT chehaoli mapinspiredartimage AT lǐzhéhào mapinspiredartimage AT chehaoli detúfēnggédeyìshùyǐngxiàng AT lǐzhéhào detúfēnggédeyìshùyǐngxiàng |
_version_ |
1718199529131474944 |