Eficient Object Pixel-Level Categorization using Bag of Features

In this paper we present a pixel-level object categorization method suitable to be applied under real-time constraints. Since pixels are categorized using a bag of features scheme, the major bottleneck of such an approach would be the feature pooling in local histograms of visual words. Therefore, we propose to bypass this time-consuming step and directly obtain the score of a linear Support Vector Machine classier. This is achieved by creating an integral image of the components of the SVM which can readily obtain the classication score for any image subwindow with only 10 additions and 2 products, regardless of its size. Besides, we evaluated the performance of two ecient feature quantization methods: the Hierarchical K-Means and the Extremely Randomized Forest. All experiments have been done in the Graz02 database, showing comparable, or even better results to related work with a lower computational cost. Source URL: https://www.iiia.csic.es/en/node/54802 Links [1] https://www.iiia.csic.es/en/staff/david-aldavert [2] https://www.iiia.csic.es/en/staff/arnau-ramisa [3] https://www.iiia.csic.es/en/staff/ricardo-toledo [4] https://www.iiia.csic.es/en/staff/ramon-l%C3%B3pez-de-m%C3%A1ntaras