Generic object recognition with boosting
This is the latest version of this eprint.
This paper presents a powerful framework for generic object recognition. Boosting is used as an underlying learning technique. We use a combination of various weak classifiers of different types of descriptors. This increases the classification result slightly but dramatically improves the stability of a classifier. Besides applying well known techniques to extract salient regions we also present a new segmentation method - ``Similarity-Measure-Segmentation''. This approach delivers segments, which can consist of several disconnected parts. This turns out to be an efficient description of local similarity. With regard to the task of object categorization, imilarity-Measure-Segmentation performs equal or better than current state-of-the-art segmentation techniques. In contrast to previous solutions we aim at handling of complex objects appearing in highly cluttered images. Therefore we have set up a database containing images with the required complexity. On these images we obtain very good classification results of up to 81% ROC-equal error rate. Focusing the performance on common databases forobject recognition our approach outperforms all comparable solutions.
Available Versions of this Item