2016 Fiscal Year Final Research Report
Query Prediction for Query-by-Sketch Image Retrieval
| Project/Area Number |
26330135
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Multimedia database
|
|---|
| Research Institution | Shizuoka University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Keywords | 画像検索 / クラスタリング / LSH / ハッシュ / SVM |
|---|
| Outline of Final Research Achievements |
The purpose of this study is to improve the efficiency of content based image retrieval by sketch. It is possible for this method to improve efficiency by the accuracy and the diversity of clustering. The proposed method is based on one of the speed-up techniques for similar image retrieval is Locality Sensitive Hashing method. This method realize to cluster features using discriminant function of Support Vector Machine as hash function and realize to keep effectiveness and improve efficiency.
|
| Free Research Field |
画像情報処理
|
