Tatiana Tommasi , KU Leuven, Leuven, ESAT-PSI and iMinds, Belgium
Francesco Orabona , Toyota Technological Institute at Chicago, Chicago
Barbara Caputo , University of Rome La Sapienza, Rome
Learning a visual object category from few samples is a compelling and challenging problem. In several real-world applications collecting many annotated data is costly and not always possible. However a small training set does not allow to cover the high intraclass variability typical of visual objects. In this condition, machine learning methods provide very few guarantees. This paper presents a discriminative model adaptation algorithm able to proficiently learn a target object with few examples by relying on other previously learned source categories. The proposed method autonomously chooses from where and how much to transfer information by solving a convex optimization problem which ensures to have the minimal leave-one-out error on the available training set. We analyze several properties of the described approach and perform an extensive experimental comparison with other existing transfer solutions, consistently showing the value of our algorithm.
Learning, Object recognition, Machine learning, Classifier design and evaluation, Computer vision
T. Tommasi, F. Orabona and B. Caputo, "Learning Categories from Few Examples with Multi Model Knowledge Transfer," in IEEE Transactions on Pattern Analysis & Machine Intelligence.