Presentation | 2008-11-28 Generating method of visual word for generic object recognition Tomoki Nishimura, HAIYUAN Wu, Toshikazu WADA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Generic object recognition aims at enabling a computer to recognize objects on non-constraint general images with their category names, which is one of the ultimate problems in computer vision research. Current main techniques in this field are generation of local feature (Bag of Feature or BoF)+construction of understanding model (pLSA, SVM,…etc). To improve the recognition accuracy, many new methods for understanding model construction have been proposed. In respect to local feature extraction that is commonly used in many recognition methods, The center points of feature are detected by using random selection or edge points or grid, and their scale information are given by random. Then the SIFT feature vectors are calculated by SIFT feature descriptors. However, there is not indemnity to conform neighboring scale on the feature points, and this may affect the recognition result adversely. In this paper, we study the method to determine the center and the scale of loca I feature that are appropriate to generic object recognition. The effectiveness of the proposed technique was confirmed by using the common data base. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | General object recognition / detection of feature vector / SVM / pLSA / SIFT feature vector |
Paper # | PRMU2008-143,MVE2008-92 |
Date of Issue |
Conference Information | |
Committee | MVE |
---|---|
Conference Date | 2008/11/20(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 | Media Experience and Virtual Environment (MVE) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Generating method of visual word for generic object recognition |
Sub Title (in English) | |
Keyword(1) | General object recognition |
Keyword(2) | detection of feature vector |
Keyword(3) | SVM |
Keyword(4) | pLSA |
Keyword(5) | SIFT feature vector |
1st Author's Name | Tomoki Nishimura |
1st Author's Affiliation | Wakayama University system department of engineering() |
2nd Author's Name | HAIYUAN Wu |
2nd Author's Affiliation | Wakayama University system department of engineering |
3rd Author's Name | Toshikazu WADA |
3rd Author's Affiliation | Wakayama University system department of engineering |
Date | 2008-11-28 |
Paper # | PRMU2008-143,MVE2008-92 |
Volume (vol) | vol.108 |
Number (no) | 328 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |