1. Active learning and discovery of object categories in the presence of unnameable instances
- Author
-
Erik Rodner, Joachim Denzler, Alexander Freytag, Paul Bodesheim, and Christoph Käding
- Subjects
Information retrieval ,business.industry ,Computer science ,Active learning (machine learning) ,Active learning ,Artificial intelligence ,Instance-based learning ,business ,Machine learning ,computer.software_genre ,Object (computer science) ,computer ,Image (mathematics) - Abstract
Current visual recognition algorithms are “hungry” for data but massive annotation is extremely costly. Therefore, active learning algorithms are required that reduce labeling efforts to a minimum by selecting examples that are most valuable for labeling. In active learning, all categories occurring in collected data are usually assumed to be known in advance and experts should be able to label every requested instance. But do these assumptions really hold in practice? Could you name all categories in every image?
- Published
- 2015
- Full Text
- View/download PDF