Comparing View-Based and Map-Based Semantic Labelling in Real-Time SLAM
This work addresses a gap in Spatial AI for researchers by providing a systematic comparison method, though it is incremental as it focuses on benchmarking rather than new techniques.
The paper tackled the lack of quantitative comparison between view-based and map-based semantic labeling in real-time SLAM, presenting an experimental framework using height map fusion to enable fair evaluation.
Generally capable Spatial AI systems must build persistent scene representations where geometric models are combined with meaningful semantic labels. The many approaches to labelling scenes can be divided into two clear groups: view-based which estimate labels from the input view-wise data and then incrementally fuse them into the scene model as it is built; and map-based which label the generated scene model. However, there has so far been no attempt to quantitatively compare view-based and map-based labelling. Here, we present an experimental framework and comparison which uses real-time height map fusion as an accessible platform for a fair comparison, opening up the route to further systematic research in this area.