| 大会名称 |
| 2016年 総合大会 |
| 大会コ－ド |
| 2016G |
| 開催年 |
|
2016
|
| 発行日 |
|
2016/3/1
|
| セッション番号 |
|
H-2
|
| セッション名 |
|
ヒューマン情報処理
|
| 講演日 |
|
2016/3/15
|
| 講演場所(会議室等) |
|
センター1号館 2F 1209
|
| 講演番号 |
|
H-2-1
|
| タイトル |
|
Preliminary Experimental Result on Image Similarity Estimation based on Eye Information
|
| 著者名 |
|
○Mario Saputra, Mutsumi Suganuma, Wataru Kameyama,
|
| キーワード |
|
image retrieval, subjective image similarity, eye movement, pupil size, eye fixation
|
| 抄録 |
|
The purpose of this research is to produce a similarity map reflecting user subjectivity using eye movement. An experiment of comparing images has been carried out. Then, image ranking was made using pairwise comparison method. Pupil size and fixation from subjects are collected using an eye tracker. Seven eye related features has been collected from five subjects. From eye features, PCA and k-means clustering algorithm are applied to get the similarity map. By comparing image subjectivity ranking and the map, we have found that some images of near rank also have near distance and grouped into the some cluster on the map.
|
| 本文pdf |
|
PDF download
|
