ICRA 2012 WORKSHOP
Semantic Perception and Mapping for Knowledge-enabled Service Robotics (with interactive session and demonstrations)
ICRA 2012, St. Paul, Minnesota, USA (Room #9, River Centre)
May 18th, 2012
9.00 - 9.30: Andrzej Pronobis, KTH, Stockholm - Introduction [ slides ]
9.30 - 10.00: Moritz Tenorth, TUM, Munich - Knowledge-enabled mapping [ slides ]
10.00 - 10.30: Dieter Fox, UW, Seatle - Grounding Natural Language [slides]
10.30 - 11.00: Coffee
11.00 - 11.30: Martial Hebert, CMU, Pittsburgh - Discovering Object Instances from Scenes of Daily Living
11.30 - 12.30: Posters and Demos session ( Full List with PDFs) [ Proceedings ]
12.30 - 14.00: Lunch
14:00 - 14.30: Richard Newcombe, Imperial College, London - Dense 3D models [slides]
14.30 - 15.00: Markus Vincze, TU Vienna, Object Classification in Domestic Environments [slides]
15.00 - 16.00: Discussion, Wrap-up, Post-discussion in the pub
====== Important Dates ======
* Submission of Posters and Demos Due: 15 April 2012
* Notification of Acceptance: 27 April 2012
* Final Submissions Due: 13 May 2012
* Workshop at ICRA 2012: 18 May 2012
====== Motivation and objectives ======
Consider a robot that is to act as a household assistant in an unknown kitchen environment. This robot has to acquire and use knowledge about where the task-relevant objects, such as the dishwasher and the oven are and how the robot can act on them. A recent advent of smart devices (e.g. smart phones) and high-quality-low-cost sensors (e.g. Kinect, Asus) provides us with the affordable resources for the robot which link sensory information to the robot's knowledge base and high-level deliberative components. Resources like this allow the general-purpose service robots to e.g. query information from world wide web, seek help from remote experts through shared autonomy interfaces and to act independently and safely in human living environments.
In this hands-on workshop we will identify key problems and solutions by narrowing down the definition of semantics, we will discuss what is the representative end world model as a result of semantic mapping, single out the optimal sensors, consider static vs. dynamic aspects of environment modeling and finally address the life-long learning in order to leverage not only the sensor data but also from human living patterns and behaviors. The workshop will feature excellent talks from researchers from academia as well as industry, live demonstrations, poster session and a working session with an aim to standardize some fundamental concepts in semantic mapping.
More specifically, we will focus on the topics related to the following:
* Over the lifetime of its operation, how should a robot's semantic mapping system build and maintain its knowledge about its surroundings?
* What are the current obstacles to bring semantic mapping capabilities to a mass-market robot?
* Which semantic mapping capabilities a robot shipped into homes today would require?
* Should we have one general system to handle many robotic tasks or should robotic systems be made for specific tasks?
====== Submissions ======
We solicit one-page A4 paper abstract submissions (formatted according to the IEEE ICRA style) for the two types of presentations: posters and live demonstrations, optionally accompanied by a video. Both will be reviewed (not double-blind) by the program committee. We would especially like to encourage live demos either during the poster session or in a special demo session and plan to foster the discussion between the participants in a plenary discussion. We plan to secure one PR2 robot to be available during the demo session. Accepted papers and videos will be archived at the Internet Archive. The authors of the best submissions will be encouraged to extend their work to the special issue of the Journal of Visual Communication and Image Representation titled Visual Understanding and Applications with RGB-D Cameras.
This is the second of two workshops at ICRA 2012 dealing with semantic perception. The earlier workshop will focus on lower-level perception, with an emphasis on RGBD sensors and 3D data, while this one will deal with higher-level perception and knowledge representation.
Topics of interest include (but are not limited to):
* Probabilistic environment modeling and representation (e.g. ontologies)
* Exploration strategies for semantic mapping and knowledge acquisition
* Strategies for active visual object search
* Semantics for long-term operation in dynamic environments
* Perceptual grounding for robots
* Classification and understanding of scenes
* Human activity recognition
* Perceived affordances of objects of daily use
* Usage of semantic maps and users’ perspective
Submission process is closed.
====== Program Committee ======
Kurt Konolige, Kei Okada, Federico Tombari, Wolfram Burgard, Moritz Tenorth, Dov Katz, Alvaro Collett, Mike Krainin, Markus Vincze, Jeanette Bohg, Nico Blodow, Jürgen Sturm, Kevin Lai, Zoltan-Csaba Marton
====== Organizers ======
* Michael Beetz, TUM, Munich
* Patric Jensfelt, KTH, Stockholm
* Alper Aydemir, KTH, Stockholm
* Dejan Pangercic, TUM, Munich
* Ben Pitzer, Bosch RTC, Palo Alto
* Bhaskara Marthi, Willow Garage, Menlo Park