Abstract
This paper studies the one-shot and zero-shot learning prob- lems, where each ob ject category has only one training example or has no training example at all. We approach this problem by transferring knowl- edge from known categories (a.k.a source categories ) to new categories (a.k.a target categories ) via ob ject attributes. Ob ject attributes are high level descriptions of ob ject categories, such as color, texture, shape, etc. Since they represent common properties across different categories, they can be used to transfer knowledge from source categories to target cat- egories effectively. Based on this insight, we propose an attribute-based transfer learning framework in this paper. We first build a generative attribute model to learn the probabilistic distributions of image features for each attribute, which we consider as attribute priors. These attribute priors can be used to (1) classify unseen images of target categories (zero- shot learning), or (2) facilitate learning classifiers for target categories when there is only one training examples per target category (one-shot learning). We demonstrate the effectiveness of the proposed approaches using the Animal with Attributes data set and show state-of-the-art per- formance in both zero-shot and one-shot learning tests.