Bi-PointFlowNet: Bidirectional Learning for Point Cloud Based Scene Flow Estimation

Cheng, Wencan; Ko, Jong Hwan

doi:10.1007/978-3-031-19815-1_7

Cited by 29 publications

(29 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Extensive experimental results demonstrate the effectiveness and generalization capability of our method. Our method surpasses the current state-of-the-art methods Bi-PointFlowNet [3]. According to the EPE3D metric, we outperform Bi-PointFlowNet by 14.6% on the FlyingTh-ings3D [19] dataset, and by 7.6% on the KITTI [21] dataset.…”

Section: Introductionmentioning

confidence: 75%

“…The methods mentioned above only use unidirectional features, which can result in insufficient information. To address this limitation, Bi-PointFlowNet [3] introduces bidirectional flow embedding layers to ex-tract correlations both forward and backward. 3DFlow [29] proposes an all-to-all flow embedding layer that can capture distant points and combine them with backward reliability validation.…”

Section: Scene Flow Estimationmentioning

confidence: 99%

“…The whole system is in a fully-supervised manner by using multi-scale loss L S the same as [1,3,36]. L S can be expressed as:…”

Section: Lossmentioning

confidence: 99%

“…Evaluation Metrics. For fair comparison, we evaluate the scene flow by following metrics, the same as [1,3,36].…”

Section: Datasets and Evaluation Metricsmentioning

confidence: 99%

“…Between the adjacent layers, it operates the process of upsampling layer, which propagates the estimated information from coarser layer to finer layer. To balance accuracy and computational efficiency, there are several methods [3,29,30] employ coarse-to-fine structure in scene flow estimation task. Despite the success of these works, they ignore the fact that the scene flow for points of the same rigid object should conform to the same geometric transformation, i.e., rigid motion consistency.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Exploiting Implicit Rigidity Constraints via Weight-Sharing Aggregation for Scene Flow Estimation from Point Clouds

Wang¹,

Cheng²,

Yang³

2023

Preprint

View full text Add to dashboard Cite

Scene flow estimation, which predicts the 3D motion of scene points from point clouds, is a core task in autonomous driving and many other 3D vision applications. Existing methods either suffer from structure distortion due to ignorance of rigid motion consistency or require explicit pose estimation and 3D object segmentation. Errors of estimated poses and segmented objects would yield inaccurate rigidity constraints and in turn mislead scene flow estimation. In this paper, we propose a novel weightsharing aggregation (WSA) method for feature and scene flow up-sampling. WSA does not rely on estimated poses and segmented objects, and can implicitly enforce rigidity constraints to avoid structure distortion in scene flow estimation. To further exploit geometric information and preserve local structure, we design a deformation degree module aim to keep the local region invariance. We modify the PointPWC-Net and integrate the proposed WSA and deformation degree module into the enhanced PointPWC-Net to derive an end-to-end scene flow estimation network, called WSAFlowNet. Extensive experimental results on the FlyingThings3D [19] and KITTI [21] datasets demonstrate that our WSAFlowNet achieves the state-of-the-art performance and outperforms previous methods by a large margin. We will release the source code at https:// github.com/wangyunlhr/WSAFlowNet.git † Equal contribution.

show abstract

Section: Introductionmentioning

confidence: 75%

Section: Scene Flow Estimationmentioning

confidence: 99%

“…The whole system is in a fully-supervised manner by using multi-scale loss L S the same as [1,3,36]. L S can be expressed as:…”

Section: Lossmentioning

confidence: 99%

“…Evaluation Metrics. For fair comparison, we evaluate the scene flow by following metrics, the same as [1,3,36].…”

Section: Datasets and Evaluation Metricsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Exploiting Implicit Rigidity Constraints via Weight-Sharing Aggregation for Scene Flow Estimation from Point Clouds

Wang¹,

Cheng²,

Yang³

2023

Preprint

View full text Add to dashboard Cite

show abstract

GMA3D: Local-Global Attention Learning to Estimate Occluded Motions of Scene Flow

Lu,

Cheng

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

RMS-FlowNet++: Efficient and Robust Multi-scale Scene Flow Estimation for Large-Scale Point Clouds

Battrawy,

Schuster,

Stricker

2024

Int J Comput Vis

View full text Add to dashboard Cite

The proposed RMS-FlowNet++ is a novel end-to-end learning-based architecture for accurate and efficient scene flow estimation that can operate on high-density point clouds. For hierarchical scene flow estimation, existing methods rely on expensive Farthest-Point-Sampling (FPS) to sample the scenes, must find large correspondence sets across the consecutive frames and/or must search for correspondences at a full input resolution. While this can improve the accuracy, it reduces the overall efficiency of these methods and limits their ability to handle large numbers of points due to memory requirements. In contrast to these methods, our architecture is based on an efficient design for hierarchical prediction of multi-scale scene flow. To this end, we develop a special flow embedding block that has two advantages over the current methods: First, a smaller correspondence set is used, and second, the use of Random-Sampling (RS) is possible. In addition, our architecture does not need to search for correspondences at a full input resolution. Exhibiting high accuracy, our RMS-FlowNet++ provides a faster prediction than state-of-the-art methods, avoids high memory requirements and enables efficient scene flow on dense point clouds of more than 250K points at once. Our comprehensive experiments verify the accuracy of RMS-FlowNet++ on the established FlyingThings3D data set with different point cloud densities and validate our design choices. Furthermore, we demonstrate that our model has a competitive ability to generalize to the real-world scenes of the KITTI data set without fine-tuning.

show abstract

Bi-PointFlowNet: Bidirectional Learning for Point Cloud Based Scene Flow Estimation

Cited by 29 publications

References 39 publications

Exploiting Implicit Rigidity Constraints via Weight-Sharing Aggregation for Scene Flow Estimation from Point Clouds

Exploiting Implicit Rigidity Constraints via Weight-Sharing Aggregation for Scene Flow Estimation from Point Clouds

GMA3D: Local-Global Attention Learning to Estimate Occluded Motions of Scene Flow

RMS-FlowNet++: Efficient and Robust Multi-scale Scene Flow Estimation for Large-Scale Point Clouds

Contact Info

Product

Resources

About