Dynamic 3D Scene Analysis by Point Cloud Accumulation

TL;DR

This paper presents a method for accumulating multi-frame 3D point clouds from LiDAR scans to create denser, more complete scene representations, improving alignment accuracy and benefiting tasks like surface reconstruction in outdoor environments.

Contribution

It introduces a novel approach that exploits scene geometry and object rigidity to improve point cloud alignment across frames, addressing challenges posed by moving objects.

Findings

Reduces alignment errors on benchmark datasets

Enhances scene coverage and density

Improves surface reconstruction quality

Abstract

Multi-beam LiDAR sensors, as used on autonomous vehicles and mobile robots, acquire sequences of 3D range scans ("frames"). Each frame covers the scene sparsely, due to limited angular scanning resolution and occlusion. The sparsity restricts the performance of downstream processes like semantic segmentation or surface reconstruction. Luckily, when the sensor moves, frames are captured from a sequence of different viewpoints. This provides complementary information and, when accumulated in a common scene coordinate frame, yields a denser sampling and a more complete coverage of the underlying 3D scene. However, often the scanned scenes contain moving objects. Points on those objects are not correctly aligned by just undoing the scanner's ego-motion. In the present paper, we explore multi-frame point cloud accumulation as a mid-level representation of 3D scan sequences, and develop a method that exploits inductive biases of outdoor street scenes, including their geometric layout and object-level rigidity. Compared to state-of-the-art scene flow estimators, our proposed approach aims to align all 3D points in a common reference frame correctly accumulating the points on the individual objects. Our approach greatly reduces the alignment errors on several benchmark datasets. Moreover, the accumulated point clouds benefit high-level tasks like surface reconstruction.