Where Next in Object Recognition and how much Supervision Do We Need?
- Sandra EbertAffiliated withMax Planck Institute for Informatics Email author
- , Bernt SchieleAffiliated withMax Planck Institute for Informatics
Object class recognition is an active topic in computer vision still presenting many challenges. In most approaches, this task is addressed by supervised learning algorithms that need a large quantity of labels to perform well. This leads either to small datasets (<10,000 images) that capture only a subset of the real-world class distribution (but with a controlled and verified labeling procedure), or to large datasets that are more representative but also add more label noise. Therefore, semi-supervised learning has been established as a promising direction to address object recognition. It requires only few labels while simultaneously making use of the vast amount of images available today. In this chapter, we outline the main challenges of semi-supervised object recognition, we review existing approaches, and we emphasize open issues that should be addressed next to advance this research topic.
- Where Next in Object Recognition and how much Supervision Do We Need?
- Book Title
- Advanced Topics in Computer Vision
- pp 35-64
- Print ISBN
- Online ISBN
- Series Title
- Advances in Computer Vision and Pattern Recognition
- Series ISSN
- Springer London
- Copyright Holder
- Springer-Verlag London
- Additional Links
- Industry Sectors
- eBook Packages
- Editor Affiliations
- 3. Dipartimento di Matematica e Informatica, Università di Catania
- 4. Dipartimento di Matematica e Informatica, Università di Catania
- 5. Department of Engineering, University of Cambridge
- Author Affiliations
- 6. Max Planck Institute for Informatics, Saarbrücken, Germany
To view the rest of this content please follow the download PDF link above.