| Project/Area Number |
16K12459
|
|---|
| Research Category |
Grant-in-Aid for Challenging Exploratory Research
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Perceptual information processing
|
|---|
| Research Institution | University of Yamanashi |
|---|
Principal Investigator |
MAO xiaoyang 山梨大学, 大学院総合研究部, 教授 (20283195)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
藤代 一成 慶應義塾大学, 理工学部(矢上), 教授 (00181347)
豊浦 正広 山梨大学, 大学院総合研究部, 准教授 (80550780)
|
|---|
| Research Collaborator |
GYOBA Jiro 東北大学, 文学研究科, 教授 (50142899)
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥3,510,000 (Direct Cost: ¥2,700,000、Indirect Cost: ¥810,000)
Fiscal Year 2018: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2017: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2016: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
|
|---|
| Keywords | 顕著性マップ |
|---|
| Outline of Final Research Achievements |
Researchers have proposed a wide variety of saliency models,ranging from models that use local,low-level image features to recent approaches that incorporate semantic information and deep learning. However, these models do not account for the visual attention related to certain global structures evident in images. We focused specifically on “leading-line” structures,in which lines converge on a single point.We have conducted the experiments to investigate the visual attentions in images with leading line structure and proposed a new saliency model combining the low level feature of center-surrounding differences of visual stimuli, the semantic feature of center-bias and the structure feature of leading lines. Experimental results show that our model outperforms the existing models with all the representative evaluation metrics of saliency map for the images with leading line structures.
|
| Academic Significance and Societal Importance of the Research Achievements |
画像提示の初期段階における人間の注意を予測する顕著性マップはロボットビジョンや広告デザインをはじめ，様々な分野で注目されている．局所的な低レベルの画像特徴を利用したモデルに加え，近年では意味論的な情報を考慮したものやDeep Learning を利用したものなど，様々な顕著性モデルが提案された．一方で，これらのモデルは画像内に存在する大域的な構造による誘目効果を考慮していないなかった．本研究の成果はロボットビジョン，広告デザイン，画像編集などへの応用が期待できるのみでなく，認知心理学分野においても多くの後続研究を誘発するポテンシャルの高さを有している．
|
