Representing shape with a spatial pyramid kernel
Anna Bosch, Andrew Zisserman, Xavier Muñoz · ACM International Conference on Image and Video Retrieval (CIVR) · 2007
The objective of this paper is classifying images by the object categories they contain, for example motorbikes or dolphins. There are three areas of novelty. First, we introduce a descriptor that represents local image shape and its spatial layout, together with a spatial pyramid kernel. These are designed so that the shape correspondence between two images can be measured by the distance between their descriptors using the kernel. Second, we generalize the spatial pyramid kernel, and learn its level weighting parameters (on a validation set). This significantly improves classification performance. Third, we show that shape and appearance kernels may be combined (again by learning parameters on a validation set).
Cited by 1,399 works. doi:10.1145/1282280.1282340
Authors
References
- Distinctive Image Features from Scale-Invariant Keypoints
- Histograms of Oriented Gradients for Human Detection
- LIBSVM
- Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories
- Libsvm : A library for support vector machines
- Learning Generative Visual Models from Few Training Examples: An Incremental Bayesian Approach Tested on 101 Object Categories
- Visual categorization with bags of keypoints
- The pyramid match kernel: discriminative classification with sets of image features
- Local Features and Kernels for Classification of Texture and Object Categories: A Comprehensive Study
- Real-time object detection for "smart" vehicles