SCIM: Simultaneous Clustering, Inference, and Mapping for Open-World Semantic Scene Understanding
This addresses the challenge of open-world semantic scene understanding for robots operating in human environments, representing an incremental advancement over existing methods.
The paper tackles the problem of enabling robots to autonomously discover novel semantic classes and improve accuracy on known classes in open-world environments by developing a framework for mapping and clustering that generates a self-supervised learning signal. The result includes optimized clustering parameters during deployment and improved novel object discovery through multi-modal fusion compared to prior work.
In order to operate in human environments, a robot's semantic perception has to overcome open-world challenges such as novel objects and domain gaps. Autonomous deployment to such environments therefore requires robots to update their knowledge and learn without supervision. We investigate how a robot can autonomously discover novel semantic classes and improve accuracy on known classes when exploring an unknown environment. To this end, we develop a general framework for mapping and clustering that we then use to generate a self-supervised learning signal to update a semantic segmentation model. In particular, we show how clustering parameters can be optimized during deployment and that fusion of multiple observation modalities improves novel object discovery compared to prior work. Models, data, and implementations can be found at https://github.com/hermannsblum/scim