Active Semantic Perception and Object Search in the Real World (ASP-AVS-11)


September 26th, 2011, San Francisco, California, in conjunction with the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)


Abstract

This workshop brings together a number of key researchers as invited speakers to discuss the problem of efficiently searching for objects in the real world. This is an important component in most service robot systems. It is a very complex problem incorporating a number of, in themselves very hard, sub-problems such as object detection, recognition and classification, view planning, active perception, mapping, spatial reasoning and more recently the incorporation and exploitation of semantic knowledge. Together with the other presenters we will give an overview of the field and discuss, together with the audience, where we are today, what the challenges and opportunities are and where the field is going.

Motivation and Objectives

A key competence for an autonomous mobile service robot operating in an unstructured indoor environment is to reliably find objects. A further requirement in this aim is the ability to acquire the relevant semantic information, update it during the whole lifetime of a robot and handle search queries in large environments. The underlying sub-problems, such as semantic perception and mapping, object recognition and categorization are the main focus of various independent research streams. Since perception algorithms can be computationally expensive, the robot must also plan its sensory actions to find the object efficiently. Our goal is to bridge the gap between those topics to allow for building robots actively seeking and integrating spatial knowledge for the purpose of efficient object search.

A common assumption present in previous work in the field is that the objects of interest are located within the immediate sensory reach of a robot. Searching for objects with mobile robots over larger regions of space has received more interest recently with various methods being developed simultaneously by different research groups. The main focus of this workshop is on autonomously finding objects in larger-scale space and methods for building a semantic spatial representation. The workshop will provide a common forum where researchers can exchange ideas and experiences with these problems.

Topics

Topics of interest include, but are not necessary limited to:
- Semantic mapping and scene understanding
- Object recognition and categorization in complex scenes
- Object search using vision or other sensory modalities
- Representations capturing semantics of large-scale environments
- Utilizing 3D cues semantic scene perception
- Planning for active perception and exploration

Setup

The workshop will consist of a mix of invited speakers, 2 or 3 paper presentations and a poster session. All submissions should be in PDF format using the IROS template. Papers should be 4-6 pages in length, optionally accompanied by a video. All submissions will be reviewed for relevance to the topic of the workshop as well as scientific merit. Submissions should be emailed to: asp-avs-11@csc.kth.se

Invited speakers

Important Dates

Program

8:30 - 8:40 - Introduction to the workshop and the problem
          Alper Aydemir, Andrzej Pronobis, Royal Institute of Technology (KTH), Sweden

8:40 - 9:05 - Invited talk
          John Tsotsos, York University, Canada

9:05 - 9:30 - Invited talk
          Jim Little, UBC, Canada

9:30 - 9:55 - Invited talk
          Charlie Kemp, Georgia Tech, USA

9:55 - 10:20 - Invited talk
          Michael Beetz, TUM, Germany

10:20 - 10:35 - Coffee break

10:35 - 11:00 - Invited talk
          Wolfram Burgard, University of Freiburg, Germany

11:00 - 12:00 - Paper session
          11:00 - 11:20 - Detecting and Localising Multiple 3D Objects: A Fast and Scalable Approach
          Dima Damen, Andrew Gee, Andrew Calway and Walterio Mayol-Cuevas

          11:20 - 11:40 - Robust Semantic Place Recognition with Vocabulary Tree and Landmark Detection
          Lin Yuan, Kai Chi Chan and C.S. George Lee

          11:40 - 12:00 - Semantic Object Search in Large-scale Indoor Environments
          Manabu Saito, Haseru Chen, Kei Okada, Masayuki Inaba, Lars Kunze, Michael Beetz

12:00 - 12:40 - Poster session

12:40 - 13:00 - Discussion, closing remarks

Program committee

Organizers