MOVES: Movable and Moving LiDAR Scene Segmentation in Label-Free settings using Static Reconstruction

TL;DR

MOVES introduces a GAN-based model that reconstructs static scenes from dynamic LiDAR scans without requiring prior segmentation, enabling better autonomous navigation in label-free settings.

Contribution

It presents a novel adversarial approach to transform dynamic LiDAR scans into static ones using static-dynamic LiDAR pairs, without relying on segmentation.

Findings

Successfully segments moving and movable objects without segmentation info

Accurately reconstructs static scenes from dynamic LiDAR data

Improves scene understanding for autonomous navigation

Abstract

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.