Our workshop will discuss the current progress, applications, and limitations of robotic scene understanding and semantic simultaneous localization and mapping (SLAM). We are motivated by the quickly accelerating appearance new research outputs that investigate how classical SLAM techniques and deep-learning based visual object detection or segmentation can be combined in innovative ways, and used to support scene understanding, navigation, and manipulation.
In addition, the workshop will host a new research challenge and competition: The Robotic Vision Scene Understanding Challenge evaluates how well a robotic vision system can understand the semantic and geometric aspects of its environment.
Call for Papers
We invite authors to submit contributed papers to the workshop. The topics of interest comprise, but are not limited to:
- new approaches of semantic SLAM and object-based SLAM
- representations of objects as landmarks in SLAM
- Deep learning for semantic SLAM & semantic SLAM for deep learning
- End-to-end and explicit scene understanding
- Incorporating high-level geometric constraints into SLAM
- Learning and applying object shape priors
- Incorporating uncertainty techniques and Bayesian deep learning into scene understanding
- Dynamic SLAM and scene understanding in non-static scenes aided by semantics
- applications of semantic SLAM and scene understanding
- success stories and failure cases
- Submissions should follow the ICRA format and can be up to 6 pages long, plus unlimited space for references.
- Please submit your paper through CMT.
- All accepted papers will be presented at a poster session.
- Selected paper will be invited for an oral presentation.
Call for Participation in the Semantic SLAM Challenge
We will organise a challenge and competition for semantic SLAM and scene understanding in conjunction with the workshop. More information coming soon. Meanwhile, the video below provides an overview of what to expect.
More information coming soon.
Confirmed invited speakers comprise Andrew Davison, Dieter Fox, Stefanie Tellex and Cesar Cadena.
The Robotic Vision Challenges organisers are with the Australian Centre for Robotic Vision at Queensland University of Technology (QUT), Monash University, the University of Adelaide, and Google AI.