Traditional supervised visual learning simply asks annotators “what” label an image should have. We propose an approach for image classification problems requiring subjective...
e by placing terms in an abstract ‘information space’ based on their occurrences in text corpora, and then allowing a user to visualize local regions of this information space....
We address the problem of weakly supervised semantic segmentation. The training images are labeled only by the classes they contain, not by their location in the image. On test im...
Alexander Vezhnevets, Vittorio Ferrari, Joachim M....
We introduce a new dissimilarity function for ranked lists, the expected weighted Hoeffding distance, that has several advantages over current dissimilarity measures for ranked s...
Most studies in statistical or machine learning based authorship attribution focus on two or a few authors. This leads to an overestimation of the importance of the features extra...