Material Detail

Autonomously Adapting Range Data Patterns for Object Detection

Autonomously Adapting Range Data Patterns for Object Detection

This video was recorded at Video Journal of Machine Learning Abstracts - Volume 4. We present a novel approach to recognizing patterns in laser range data that performs on a par with the state of the art while at the same requiring minimal parameters and supervision. Most importantly, supervision is only needed at the level of real-world objects that a robot can interact with (humans, in our experiments). This is an important step towards autonomous cognitive systems, since the system can interact with such objects and autonomously collect all the supervision it needs in order to adapt its models.

Quality

  • User Rating
  • Comments
  • Learning Exercises
  • Bookmark Collections
  • Course ePortfolios
  • Accessibility Info

More about this material

Comments

Log in to participate in the discussions or sign up if you are not already a MERLOT member.