CFP
OSRPW2024: Open-Set Robot Perception in the Wild Delft University of Technology Delft, Netherlands, July 15-19, 2024 |
Conference website | https://open-set-robotics.github.io/ |
Submission link | https://easychair.org/conferences/?conf=osrpw2024 |
Abstract registration deadline | May 19, 2024 |
Submission deadline | May 19, 2024 |
We invite you to contribute a short paper or an extended abstract to the upcoming RSS 2024 Workshop on "Open-Set Robot Perception in the Wild." This call is for 2-4 pages of innovative research papers, with no limits on the number of references or supplementary materials. All submissions should follow the RSS 2024 paper format guidelines (LaTeX or Word).
Our workshop aims to delve into the open-set perception, a common problem in robotic systems' ability to understand and interact with the dynamically changing and unpredictable real world. We are particularly interested in contributions that address, but are not limited to, the following thematic areas:
Detection of Open-Set Samples:
- Strategies for identifying out-of-distribution, anomalous, or novel entities through detection and segmentation techniques.
- Approaches to enhance machine vision with robust uncertainty estimation.
- Exploration of the open-set limitations inherent to foundation models and proposing avenues for overcoming these challenges.
Learning from Open-Set Samples:
- Novel methodologies for applying transfer learning and domain adaptation to improve robot perception.
- Cutting-edge research on continual learning and interactive learning strategies tailored for robotic applications.
- Innovations in self-supervised learning within real-world deployments.
- Critical analyses and solutions aimed at ensuring fairness, transparency, and privacy within adaptive perception systems.