SpSequenceNet: Semantic Segmentation Network on 4D Point Clouds

Shi, Hanyu; Lin, Guosheng; Wang, Hao; Hung, Tzu-Yi; Wang, Zhenhua

doi:10.1109/cvpr42600.2020.00463

Cited by 84 publications

(43 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3. It is motivated by the Cross-frame Global Attention (CGA) in [34], where the attention mask for the current frame is generated from the information in the last frame. In our work, since our aim is to locate the target in the search area, features of the search area that match the target features should be highlighted, which we achieve through generating an attention mask.…”

Section: Target-guided Attentionmentioning

confidence: 99%

MLVSNet: Multi-level Voting Siamese Network for 3D Visual Tracking

Wang

Xie

Lai

et al. 2021

2021 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

Benefiting from the excellent performance of Siamesebased trackers, huge progress on 2D visual tracking has been achieved. However, 3D visual tracking is still underexplored. Inspired by the idea of Hough voting in 3D object detection, in this paper, we propose a Multi-level Voting Siamese Network (MLVSNet) for 3D visual tracking from outdoor point cloud sequences. To deal with sparsity in outdoor 3D point clouds, we propose to perform Hough voting on multi-level features to get more vote centers and retain more useful information, instead of voting only on the final level feature as in previous methods. We also design an efficient and lightweight Target-Guided Attention (TGA) module to transfer the target information and highlight the target points in the search area. Moreover, we propose a Vote-cluster Feature Enhancement (VFE) module to exploit the relationships between different vote clusters. Extensive experiments on the 3D tracking benchmark of KITTI dataset demonstrate that our MLVSNet outperforms state-of-the-art methods with significant margins. Code will be available at https://github.com/CodeWZT/MLVSNet.

show abstract

Section: Target-guided Attentionmentioning

confidence: 99%

MLVSNet: Multi-level Voting Siamese Network for 3D Visual Tracking

Wang

Xie

Lai

et al. 2021

2021 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

show abstract

“…the residual between LiDAR scans, free space checking, and region growing to find moving objects. There are also multiple 3D point cloud-based semantic segmentation approaches [28], [27], [26], which also perform well in semantic segmentation tasks. Among them, Shi et al [26] exploit sequential point clouds and predict moving objects.…”

Section: Related Workmentioning

confidence: 99%

Moving Object Segmentation in 3D LiDAR Data: A Learning-based Approach Exploiting Sequential Data

Chen,

Li,

Mersch

et al. 2021

Preprint

View full text Add to dashboard Cite

The ability to detect and segment moving objects in a scene is essential for building consistent maps, making future state predictions, avoiding collisions, and planning. In this paper, we address the problem of moving object segmentation from 3D LiDAR scans. We propose a novel approach that pushes the current state of the art in LiDAR-only moving object segmentation forward to provide relevant information for autonomous robots and other vehicles. Instead of segmenting the point cloud semantically, i.e., predicting the semantic classes such as vehicles, pedestrians, buildings, roads, etc., our approach accurately segments the scene into moving and static objects, i.e., distinguishing between moving cars vs. parked cars. Our proposed approach exploits sequential range images from a rotating 3D LiDAR sensor as an intermediate representation combined with a convolutional neural network and runs faster than the frame rate of the sensor. We compare our approach to several other state-of-the-art methods showing superior segmentation quality in urban environments. Additionally, we created a new benchmark for LiDAR-based moving object segmentation based on SemanticKITTI. We publish it to allow other researchers to compare their approaches transparently and we will publish our code.

show abstract

“…For the task of motion segmentation two approaches have been widely used: Networks either incorporate multiple point clouds directly or accumulate a sequence of individually segmented point clouds. Shi et al (2020) present their U-Net based architecture SpSequenceNet for semantic segmentation on 4D point clouds. They input two point clouds and generate the output for the later one with a voxel-based method.…”

Section: Motion Segmentationmentioning

confidence: 99%

LatticeNet: fast spatio-temporal point cloud segmentation using permutohedral lattices

Roşu¹,

Schütt²,

Quenzel³

et al. 2021

Auton Robot

View full text Add to dashboard Cite

Deep convolutional neural networks have shown outstanding performance in the task of semantically segmenting images. Applying the same methods on 3D data still poses challenges due to the heavy memory requirements and the lack of structured data. Here, we propose LatticeNet, a novel approach for 3D semantic segmentation, which takes raw point clouds as input. A PointNet describes the local geometry which we embed into a sparse permutohedral lattice. The lattice allows for fast convolutions while keeping a low memory footprint. Further, we introduce DeformSlice, a novel learned data-dependent interpolation for projecting lattice features back onto the point cloud. We present results of 3D segmentation on multiple datasets where our method achieves state-of-the-art performance. We also extend and evaluate our network for instance and dynamic object segmentation.

show abstract

SpSequenceNet: Semantic Segmentation Network on 4D Point Clouds

Cited by 84 publications

References 22 publications

MLVSNet: Multi-level Voting Siamese Network for 3D Visual Tracking

MLVSNet: Multi-level Voting Siamese Network for 3D Visual Tracking

Moving Object Segmentation in 3D LiDAR Data: A Learning-based Approach Exploiting Sequential Data

LatticeNet: fast spatio-temporal point cloud segmentation using permutohedral lattices

Contact Info

Product

Resources

About