Learn More
Many tasks in computer vision, such as action classification and object detection, require us to rank a set of samples according to their relevance to a particular visual category. The performance of such tasks is often measured in terms of the average precision (AP). Yet it is common practice to employ the support vector machine (SVM) classifier, which(More)
The problem of ranking a set of visual samples according to their relevance to a query plays an important role in computer vision. The traditional approach for ranking is to train a binary classifier such as a support vector machine (svm). Binary classifiers suffer from two main deficiencies: (i) they do not optimize a ranking-based loss function, for(More)
Learn parameters w t , ξ t by solving the following convex problem over the set of active constraints W, argmin w.ξ 1 2 ||w|| 2 + Cξ s.t. Add the most violated constraint to set of active constraints W. Algorithm 1 describes a cutting plane algorithm for estimating the parameters of an AP-SVM by solving the following problem: min w 1 2 ||w|| 2 + Cξ, (2)(More)
The problem of ranking a set of visual samples according to their relevance to a query plays an important role in computer vision. The traditional approach for ranking is to train a binary classifier such as a support vector machine (svm). Binary classifiers suffer from two main deficiencies: (i) they do not optimize a ranking-based loss function, for(More)
  • 1