Abstract

We present an approach to determine the category and loca- tion of ob jects in images. It performs very fast categorization of each pixel in an image, a brute-force approach made feasible by three key develop- ments: First, our method reduces the size of a large generic dictionary (on the order of ten thousand words) to the low hundreds while increas- ing classification performance compared to k-means. This is achieved by creating a discriminative dictionary tailored to the task by following the information bottleneck principle. Second, we perform feature-based cate- gorization efficiently on a dense grid by extending the concept of integral images to the computation of local histograms. Third, we compute SIFT descriptors densely in linear time. We compare our method to the state of the art and find that it excels in accuracy and simplicity, performing better while assuming less.