Loading...
Please wait, while we are loading the content...
Similar Documents
A novel, efficient, tree-based descriptor and matching algorithm (2012).
| Content Provider | CiteSeerX |
|---|---|
| Author | Fowers, Spencer G. Lee, D. J. Ventura, Dan Wilde, Doran K. |
| Abstract | This paper presents the development of a new feature descriptor derived from the BASIS descriptor that provides improvements in descriptor size, computation speed, matching speed, and accuracy. The TreeBASIS descriptor algorithm utilizes a binary vocabulary tree that is computed off-line using basis dictionary images (BDIs) derived from sparse coding and a test set of feature region images (FRIs), and the resulting tree is stored in memory for on-line searching. During the on-line algorithm, a feature region image (FRI) is binary quantized and the resulting quantized vector is passed into the BASIS tree, where a Hamming distance is computed between the FRI and the effectively descriptive BDI (EDBDI) at the current node to determine the branch taken. The path the FRI takes is saved as the descriptor, and matching is performed by following the paths of two features and iteratively reducing the distance as the path is traversed. Experimental results show that the TreeBASIS descriptor outperforms BASIS, SIFT, and SURF on frame-to-frame aerial feature point matching. 1 |
| File Format | |
| Publisher Date | 2012-01-01 |
| Access Restriction | Open |
| Subject Keyword | Tree-based Descriptor Matching Algorithm Feature Region Image Basis Tree Computation Speed Descriptive Bdi Treebasis Descriptor Algorithm Basis Dictionary Image Test Set Binary Vocabulary Tree Current Node On-line Searching Descriptor Size Quantized Vector Basis Descriptor Hamming Distance Binary Quantized On-line Algorithm Treebasis Descriptor Outperforms Basis New Feature Descriptor Frame-to-frame Aerial Feature Point Matching Sparse Coding Resulting Tree Experimental Result |
| Content Type | Text |
| Resource Type | Article |
