Automatic Labeling to Generate Training Data for Online LiDAR-Based Moving Object Segmentation

Chen, Xieyuanli; Mersch, Benedikt; Nunes, Lucas; Marcuzzi, Rodrigo; Vizzo, Ignacio; Behley, Jens; Stachniss, Cyrill

doi:10.1109/lra.2022.3166544

Cited by 62 publications

(26 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our method can outperform all baselines with an IoU MOS of 65.2%, which demonstrates the effectiveness of our approach. Our performance is also better than LMNet+AutoMOS+Extra [6], which additionally uses automatically generated moving object labels for training. This emphasizes the strength of our result.…”

Section: B Moving Object Segmentation Performancementioning

confidence: 94%

“…The back-projection from these 2D representations to the 3D space often requires post-processing like k-nearest neighbor (kNN) clustering [5], [9], [12], [25] to avoid labels bleeding into points that are close in the image space but distant in 3D. Other approaches can identify objects that have moved in 3D space directly during mapping [1] or with a clustering and tracking approach [6]. Nevertheless, these offline methods often rely on having access to all LiDAR observations in the sequence.…”

Section: Non-moving Movingmentioning

confidence: 99%

“…Other researchers encode non-static objects into the map by estimating multi-modal states [33]. Recently, Chen et al [6] propose a pipeline to automatically label moving objects offline. They first use an occupancy-based method to find dynamic point candidates and further identify moving objects by sequential clustering and tracking.…”

Section: Related Workmentioning

confidence: 99%

“…LMNet+AutoMOS+Extra [6] 62.3 TABLE I: Performance on SemanticKITTI [3] moving object segmentation benchmark [5]. Baseline results taken from [6]. Best result in bold.…”

Section: A Experimental Setupmentioning

confidence: 99%

“…To evaluate the generalization capability of our approach across environments, we additionally test it on another dataset without the use of domain adaptation techniques. We follow the setup of Chen et al [6] and use the Apollo-ColumbiaParkMapData [22] dataset sequence 2 (frames 22300-24300) and sequence 3 (frames 3100-3600) annotated the same way as SemanticKITTI. Note that Se-manticKITTI and Apollo both use Velodyne HDL-64E LiDAR scanners, but they are mounted on a different car at a different height and recorded data in a different environment.…”

Section: A Experimental Setupmentioning

confidence: 99%

See 4 more Smart Citations

Receding Moving Object Segmentation in 3D LiDAR Data Using Sparse 4D Convolutions

Mersch

Chen

Vizzo

et al. 2022

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

A key challenge for autonomous vehicles is to navigate in unseen dynamic environments. Separating moving objects from static ones is essential for navigation, pose estimation, and understanding how other traffic participants are likely to move in the near future. In this work, we tackle the problem of distinguishing 3D LiDAR points that belong to currently moving objects, like walking pedestrians or driving cars, from points that are obtained from non-moving objects, like walls but also parked cars. Our approach takes a sequence of observed LiDAR scans and turns them into a voxelized sparse 4D point cloud. We apply computationally efficient sparse 4D convolutions to jointly extract spatial and temporal features and predict moving object confidence scores for all points in the sequence. We develop a receding horizon strategy that allows us to predict moving objects online and to refine predictions on the go based on new observations. We use a binary Bayes filter to recursively integrate new predictions of a scan resulting in more robust estimation. We evaluate our approach on the SemanticKITTI moving object segmentation challenge and show more accurate predictions than existing methods. Since our approach only operates on the geometric information of point clouds over time, it generalizes well to new, unseen environments, which we evaluate on the Apollo dataset. Index Terms-Semantic Scene Understanding; Deep Learning Methods I. INTRODUCTION D ISTINGUISHING moving from static objects in 3D LiDAR data is a crucial task for autonomous systems and required for planning collision-free trajectories and navigating safely in dynamic environments. Moving object segmentation (MOS) can improve localization [5], [7], planning [34], mapping [5], scene flow estimation [2], [15], [37], or the prediction of future states [38], [40]. There are mapping approaches that identify if observed points are potentially moving or have moved throughout the mapping process [1], [7], [16], [28]. On the contrary, identifying objects that are actually moving within a short time horizon are of interest for online navigation [34], can improve scene flow estimation between two consecutive point clouds [2], [15], [37], or support predicting a future state of the environment [40].

show abstract

Section: B Moving Object Segmentation Performancementioning

confidence: 94%

Section: Non-moving Movingmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

“…LMNet+AutoMOS+Extra [6] 62.3 TABLE I: Performance on SemanticKITTI [3] moving object segmentation benchmark [5]. Baseline results taken from [6]. Best result in bold.…”

Section: A Experimental Setupmentioning

confidence: 99%

Section: A Experimental Setupmentioning

confidence: 99%

See 3 more Smart Citations