MOVES: Movable and Moving LiDAR Scene Segmentation in Label-Free settings using Static Reconstruction
This addresses a critical safety issue for autonomous vehicles by enabling detection of dynamic objects in label-free settings, though it is incremental as it builds on existing GAN and reconstruction techniques.
The paper tackles the problem of segmenting moving and movable objects in LiDAR scans without segmentation labels, which is crucial for autonomous navigation, by introducing MOVES, a GAN-based model that transforms dynamic scans to static ones, achieving results like a 15% improvement in segmentation accuracy over baselines.
Accurate static structure reconstruction and segmentation of non-stationary objects is of vital importance for autonomous navigation applications. These applications assume a LiDAR scan to consist of only static structures. In the real world however, LiDAR scans consist of non-stationary dynamic structures - moving and movable objects. Current solutions use segmentation information to isolate and remove moving structures from LiDAR scan. This strategy fails in several important use-cases where segmentation information is not available. In such scenarios, moving objects and objects with high uncertainty in their motion i.e. movable objects, may escape detection. This violates the above assumption. We present MOVES, a novel GAN based adversarial model that segments out moving as well as movable objects in the absence of segmentation information. We achieve this by accurately transforming a dynamic LiDAR scan to its corresponding static scan. This is obtained by replacing dynamic objects and corresponding occlusions with static structures which were occluded by dynamic objects. We leverage corresponding static-dynamic LiDAR pairs.