Retrieving Objects Using Local Integral Invariants
The use of local features in computer vision has shown to be promising. Local features have several advantages including invariance to image transformations, independence of the background, and robustness in difficult situations like partial occlusions. In this paper we suggest using local integral invariants to extract local image descriptors around interest points and use them for the retrieval task. Integral invariants capture the local structure of the neighborhood around the points where they are computed. This makes them very well suited for constructing highly-discriminative local descriptors. We study two types of kernels used for extracting the feature vectors and compare the performance of both. The dimensionality of the feature vector to be used is investigated. We also compare our results with the SIFT features. Excellent results are obtained using a dataset that contains instances of objects that are viewed in difficult situations that include clutter and occlusion.
KeywordsFeature Vector Recognition Rate Local Binary Pattern Interest Point Query Image
Unable to display preview. Download preview PDF.
- 2.Schiele, B., Crowley, J.L.: Object Recognition Using Multidimensional Receptive Field Histograms. In: ECCV., vol. 1, pp. 610–619 (1996) Google Scholar
- 3.Schmid, C., Mohr, R.: Local Grayvalue Invariants for Image Retrieval. PAMI 19, 530–535 (1997)Google Scholar
- 4.Harris, C., Stephens, M.: A Combined Corner and Edge Detector. In: Proceedings of the Fourth Alvey Vision Conference, pp. 147–151 (1988)Google Scholar
- 5.Gouet, V., Boujemaa, N.: Object-based Queries using Color Points of Interest. In: CBAIVL, Kauai, Hawaii, USA, pp. 30–36 (2001)Google Scholar
- 7.Schulz-Mirbach, H.: Invariant Features for Gray Scale Images. In: 17th DAGM, Bielefeld, pp. 1–14 (1995)Google Scholar
- 8.Schael, M.: Invariant Grey Scale Features for Texture Analysis Based on Group Averaging with Relational Kernel Functions. Technical Report 1/01, University of Freiburg (2001)Google Scholar
- 10.Siggelkow, S.: Feature Historgrams for Content-Based Image Retrieval. Ph.D thesis, Albert-Ludwigs-Universität, Freiburg (2002)Google Scholar
- 12.Biber, P., Straßer, W.: Solving the Correspondence Problem by Finding Unique Features. In: 16th International Conference on Vision Interface (2003)Google Scholar
- 13.Mikolajczyk, K., Schmid, C.: A Performance Evaluation of Local Descriptors. In: CVPR, vol. 2, pp. 257–263 (2003)Google Scholar