| Presentation | 2011-02-17 Automatic Image Annotation by Variational Random Forests Motofumi FUKUI, Noriji KATO, Wenyuan QI, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Recently automatic image annotation (AIA) receives a lot of attention in the fields of information retrieval, and many ideas have been proposed over this decades. Semantic Multi-Class Labeling (SML) is the representative example, based on a Naive Bayes classifier with local image features whose existing probabilities are estimated by gaussian mixture models. However there is such a difficult problem that SML requires its long training time and annotation time, when a posterior distribution of each label is estimated. To deal with this problem, instead of gaussian mixture models, we introduce a Random Forest classifier into the AIA task. We have evaluated our method by using a standard image corpus. We show that the speed of annotating by our method is about 50 times faster than by SML, with maintaining the same performance as the existing ideas. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Image Processing / Automatic Image Annotation / Image Retrieval / Random Forest / Generative Model |
| Paper # | PRMU2010-209 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2011/2/10(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Pattern Recognition and Media Understanding (PRMU) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Automatic Image Annotation by Variational Random Forests |
| Sub Title (in English) | |
| Keyword(1) | Image Processing |
| Keyword(2) | Automatic Image Annotation |
| Keyword(3) | Image Retrieval |
| Keyword(4) | Random Forest |
| Keyword(5) | Generative Model |
| 1st Author's Name | Motofumi FUKUI |
| 1st Author's Affiliation | Corporate Research & Technology Group, Fuji Xerox Co., Ltd.() |
| 2nd Author's Name | Noriji KATO |
| 2nd Author's Affiliation | Corporate Research & Technology Group, Fuji Xerox Co., Ltd. |
| 3rd Author's Name | Wenyuan QI |
| 3rd Author's Affiliation | Corporate Research & Technology Group, Fuji Xerox Co., Ltd. |
| Date | 2011-02-17 |
| Paper # | PRMU2010-209 |
| Volume (vol) | vol.110 |
| Number (no) | 414 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |