Institutionen för systemteknik (ISY)

Informed Visual Search: Combining Attention and Object Recognition

Per-Erik Forssén, David Meger, Kevin Lai, Scott Helmer, James J. Little, David G. Lowe
ICRA08, Pasadena, CA, USA
IEEE International Conference on Robotics and Automation
Volume CFP08RAA-CDR, Pages 935-942
May 2008

Abstract

This paper studies the sequential object recognition problem faced by a mobile robot searching for specific objects within a cluttered environment. In contrast to current state-of-the-art object recognition solutions which are evaluated on databases of static images, the system described in this paper employs an active strategy based on identifying potential objects using an attention mechanism and planning to obtain images of these objects from numerous viewpoints. We demonstrate the use of a bag-of-features technique for ranking potential objects, and show that this measure outperforms geometric matching for invariance across viewpoints. Our system implements informed visual search by prioritising locations and searching promising locations first. Experimental results demonstrate that our system is a highly competent object recognition system that is capable of locating numerous challenging objects amongst distractors.

Full Paper

Portable document format file PDF ()
This system won the SRVC Comptetition at the AAAI07 conference in Vancouver. [Curious George page]

---

Bibtex entry

@InProceedings{forssen08,
  author = 	 {Per-Erik Forss{\'e}n and David Meger and Kevin Lai and  Scott Helmer and James J. Little and David G. Lowe},
  title = 	 {Informed Visual Search: Combining Attention and Object Recognition},
  booktitle = {{IEEE} International Conference on Robotics and Automation},
  pages = 	 {935-942},
  year = 	 {2008},
  volume = 	 {CFP08RAA-CDR},
  address = 	 {Pasadena, CA, USA},
  month = 	 {May},
  organization = {IEEE},
  isbn =         {978-1-4244-1647-9},
  publisher =    {{IEEE} Robotics and Automation Society}
}