IRIS-SLAM: Unified Geo-Instance Representations for Robust Semantic Localization and Mapping
This work addresses the challenge of robust semantic mapping for robotics or autonomous systems, though it appears incremental as it builds on existing geometry foundation models.
The paper tackles the problem of lacking deep semantic understanding and robust loop closure in dense geometric SLAM systems by proposing IRIS-SLAM, which uses unified geometric-instance representations to improve semantic localization and mapping, resulting in significant outperformance over state-of-the-art methods in map consistency and loop closure reliability.
Geometry foundation models have significantly advanced dense geometric SLAM, yet existing systems often lack deep semantic understanding and robust loop closure capabilities. Meanwhile, contemporary semantic mapping approaches are frequently hindered by decoupled architectures and fragile data association. We propose IRIS-SLAM, a novel RGB semantic SLAM system that leverages unified geometric-instance representations derived from an instance-extended foundation model. By extending a geometry foundation model to concurrently predict dense geometry and cross-view consistent instance embeddings, we enable a semantic-synergized association mechanism and instance-guided loop closure detection. Our approach effectively utilizes viewpoint-agnostic semantic anchors to bridge the gap between geometric reconstruction and open-vocabulary mapping. Experimental results demonstrate that IRIS-SLAM significantly outperforms state-of-the-art methods, particularly in map consistency and wide-baseline loop closure reliability.