Academic
Publications
Object Class Recognition Using Multiple Layer Boosting with Heterogeneous Features

Object Class Recognition Using Multiple Layer Boosting with Heterogeneous Features,10.1109/CVPR.2005.251,Wei Zhang,Bing Yu,Gregory J. Zelinsky,Dimitri

Object Class Recognition Using Multiple Layer Boosting with Heterogeneous Features   (Citations: 38)
BibTex | RIS | RefWorks Download
We combine local texture features (PCA-SIFT), global features (shape context), and spatial features within a sin- gle multi-layer AdaBoost model of object class recognition. The first layer selects PCA-SIFT and shape context features and combines the two feature types to form a strong clas- sifier. Although previous approaches have used either fea- ture type to train an AdaBoost model, our approach is the first to combine these complementary sources of informa- tion into a single feature pool and to use Adaboost to select those features most important for class recognition. The second layer adds to these local and global descriptions in- formation about the spatial relationships between features. Through comparisons to the training sample, we first find the most prominent local features in Layer 1, then capture the spatial relationships between these features in Layer 2. Rather than discarding this spatial information, we there- fore use it to improve the strength of our classifier. We com- pared our method to (4, 12, 13) and in all cases our ap- proach outperformed these previous methods using a popu- lar benchmark for object class recognition (4). ROC equal error rates approached 99%. We also tested our method using a dataset of images that better equates the complex- ity between object and non-object images, and again found that our approach outperforms previous methods.
Conference: Computer Vision and Pattern Recognition - CVPR , vol. 2, pp. 323-330, 2005
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
Sort by: