Presentation | 2009-06-19 Matching with Local Invariant Features Based on Dense Edge Sampling Naoyuki ICHIMURA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Detecting local regions in which descriptors are computed is necessary to extract local invariant features for image matching. The filters for feature point extraction such as LoG (Laplacian of Gaussian) have been used to find the appropriate positions of the local regions in an image. In this paper, we point out on local region detection that the portions of an image with intensity variations useful for image matching are not used as the local regions due to the difference between the sizes of the filters and the local regions. In order to take full advantage of intensity variations, we propose a method to detect the local regions based on dense edge sampling. Using the entropies of descriptors, we quantitatively show that the number of local regions with intensity variations useful for image matching is greatly increased by dense edge sampling. Experimental results obtained by a GPU-based implementation demonstrate the robustness of the proposed method to scenes with occlusions and less textures. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | |
Paper # | PRMU2009-51 |
Date of Issue |
Conference Information | |
Committee | PRMU |
---|---|
Conference Date | 2009/6/11(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) | Matching with Local Invariant Features Based on Dense Edge Sampling |
Sub Title (in English) | |
Keyword(1) | |
1st Author's Name | Naoyuki ICHIMURA |
1st Author's Affiliation | Neuroscience Research Institute, National Institute of Advanced Industrial Science and Technology (AIST)() |
Date | 2009-06-19 |
Paper # | PRMU2009-51 |
Volume (vol) | vol.109 |
Number (no) | 88 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |