Search papers, labs, and topics across Lattice.
The paper introduces active mapping of moving objects, a novel task where a mapping agent plans its trajectory while accounting for object motion to improve 3D reconstruction. Paparazzo, a learning-free approach, predicts the target's trajectory and selects informative viewpoints to plan the agent's path. Experiments on a newly introduced benchmark demonstrate that Paparazzo significantly enhances 3D reconstruction completeness and accuracy compared to existing methods.
Forget static scenes: Paparazzo lets robots actively map *moving* 3D objects with unprecedented accuracy.
Current 3D mapping pipelines generally assume static environments, which limits their ability to accurately capture and reconstruct moving objects. To address this limitation, we introduce the novel task of active mapping of moving objects, in which a mapping agent must plan its trajectory while compensating for the object's motion. Our approach, Paparazzo, provides a learning-free solution that robustly predicts the target's trajectory and identifies the most informative viewpoints from which to observe it, to plan its own path. We also contribute a comprehensive benchmark designed for this new task. Through extensive experiments, we show that Paparazzo significantly improves 3D reconstruction completeness and accuracy compared to several strong baselines, marking an important step toward dynamic scene understanding. Project page: https://davidea97.github.io/paparazzo-page/
