CMU-Flownet: Exploring Point Cloud Scene Flow Estimation in Occluded Scenario (2404.10571v1)
Abstract: Occlusions hinder point cloud frame alignment in LiDAR data, a challenge inadequately addressed by scene flow models tested mainly on occlusion-free datasets. Attempts to integrate occlusion handling within networks often suffer accuracy issues due to two main limitations: a) the inadequate use of occlusion information, often merging it with flow estimation without an effective integration strategy, and b) reliance on distance-weighted upsampling that falls short in correcting occlusion-related errors. To address these challenges, we introduce the Correlation Matrix Upsampling Flownet (CMU-Flownet), incorporating an occlusion estimation module within its cost volume layer, alongside an Occlusion-aware Cost Volume (OCV) mechanism. Specifically, we propose an enhanced upsampling approach that expands the sensory field of the sampling process which integrates a Correlation Matrix designed to evaluate point-level similarity. Meanwhile, our model robustly integrates occlusion data within the context of scene flow, deploying this information strategically during the refinement phase of the flow estimation. The efficacy of this approach is demonstrated through subsequent experimental validation. Empirical assessments reveal that CMU-Flownet establishes state-of-the-art performance within the realms of occluded Flyingthings3D and KITTY datasets, surpassing previous methodologies across a majority of evaluated metrics.
- Ssflownet: Semi-supervised scene flow estimation on point clouds with pseudo label. arXiv preprint arXiv:2312.15271 .
- Bi-pointflownet: Bidirectional learning for point cloud based scene flow estimation, in: European Conference on Computer Vision, Springer. pp. 108–124.
- Hplflownet: Hierarchical permutohedral lattice flownet for scene flow estimation on large-scale point clouds. Cornell University - arXiv,Cornell University - arXiv .
- Hplflownet: Hierarchical permutohedral lattice flownet for scene flow estimation on large-scale point clouds, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 3254–3263.
- A variational method for scene flow estimation from stereo sequences, in: 2007 IEEE 11th International Conference on Computer Vision, IEEE. pp. 1–7.
- Flowstep3d: Model unrolling for self-supervised scene flow estimation. Cornell University - arXiv,Cornell University - arXiv .
- Hierarchical edge aware learning for 3d point cloud, in: Computer Graphics International Conference, Springer. pp. 81–92.
- Flownet3d: Learning scene flow in 3d point clouds, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 529–537.
- Gma3d: Local-global attention learning to estimate occluded motions of scene flow, in: Chinese Conference on Pattern Recognition and Computer Vision (PRCV), Springer. pp. 16–27.
- A large dataset to train convolutional networks for disparity, optical flow, and scene flow estimation, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 4040–4048.
- Object scene flow for autonomous vehicles, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 3061–3070.
- Object scene flow. ISPRS Journal of Photogrammetry and Remote Sensing 140, 60–76.
- Semantic video segmentation by gated recurrent flow propagation, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 6819–6828.
- Occlusion guided scene flow estimation on 3d point clouds, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2805–2814.
- Flot: Scene flow on point clouds guided by optimal transport. Cornell University - arXiv,Cornell University - arXiv .
- Pointnet: Deep learning on point sets for 3d classification and segmentation, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 652–660.
- Pointnet++: Deep hierarchical feature learning on point sets in a metric space. Advances in neural information processing systems 30.
- Pwoc-3d: Deep occlusion-aware end-to-end scene flow estimation, in: 2019 IEEE Intelligent Vehicles Symposium (IV), IEEE. pp. 324–331.
- Self-supervised 3d scene flow estimation guided by superpoints, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5271–5280.
- Point-gnn: Graph neural network for 3d object detection in a point cloud, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 1711–1719.
- Learning video object segmentation with visual memory, in: Proceedings of the IEEE International Conference on Computer Vision, pp. 4481–4490.
- Three-dimensional scene flow, in: Proceedings of the Seventh IEEE International Conference on Computer Vision, IEEE. pp. 722–729.
- 3d scene flow estimation with a rigid motion prior, in: 2011 International Conference on Computer Vision, IEEE. pp. 1291–1298.
- What matters for 3d scene flow network, in: European Conference on Computer Vision, Springer. pp. 38–55.
- Hierarchical attention learning of scene flow in 3d point clouds. IEEE Transactions on Image Processing 30, 5168–5181.
- Festa: Flow estimation via spatial-temporal attention for scene point clouds, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 14173–14182.
- Estimation and propagation: Scene flow prediction on occluded point clouds. IEEE Robotics and Automation Letters 7, 12201–12208.
- Pv-raft: Point-voxel correlation fields for scene flow estimation of point clouds, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 6954–6963.
- Pointpwc-net: A coarse-to-fine network for supervised and self-supervised scene flow estimation on 3d point clouds. Cornell University - arXiv,Cornell University - arXiv .
- Learning scene flow from 3d point clouds with cross-transformer and global motion cues, in: ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE. pp. 1–5.
- Maskflownet: Asymmetric feature matching with learnable occlusion mask, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 6278–6287.
Sponsor
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.