As substantial progress has been made in the segmentation of 2D images using convolutional neural networks [14, 19, 24], interest in the problem of 3D semantic segmentation has grown recently. This interest was fueled, in particular, by the release of a new dataset for the part-based segmentation of 3D objects, and an associated competition.. We demonstrate the strong performance of the resulting mod-els, called submanifold sparse convolutional networks (SS-CNs), on two tasks involving semantic segmentation of 3D point clouds. In particular, our models outperform all prior state-of-the-art on the test set of a recent semantic segmen-tation competition. 1.

论文阅读：基于事件的异步稀疏卷积（ECCV 2020） 知乎

论文阅读：基于事件的异步稀疏卷积（ECCV 2020） 知乎

论文阅读：3D Semantic Segmentation with Submanifold Sparse Convolutional

Figure 1 from OccuSeg Occupancyaware 3D Instance Segmentation

[PDF] Submanifold Sparse Convolutional Networks Semantic Scholar

3D Semantic Segmentation

Submanifold Sparse Convolutional Networks

(PDF) SUBMANIFOLD SPARSE CONVOLUTIONAL NETWORKS FOR SEMANTIC

Submanifold Sparse Convolutional Networks 知乎

稀疏卷积【1】Submanifold Sparse Convolutional NetworksCSDN博客

3D Classification AI牛丝

(PDF) Semantic Segmentation with a Sparse Convolutional Neural Network

(PDF) Automated Segmentation of Computed Tomography Images with

阅读笔记[CVPR2018] 3D Semantic Segmentation with Submanifold Sparse

3D Semantic Segmentation with Submanifold Sparse Convolutional Networks

Efficient Neighbourhood Consensus Networks via Submanifold Sparse

Figure 1 from Cosegmentation of Textured 3D Shapes with Sparse

论文阅读：3D Semantic Segmentation with Submanifold Sparse Convolutional

[PDF] 3D Graph Embedding Learning with a Structureaware Loss Function

deep learning convolutional neural network Cross Validated

Submanifold Sparse Convolutions 3D Semantic Segmentation with Submanifold Sparse Convolutional Networks (arxiv: 1711.10275) Dilation problem 1 nonzero site leads to 3d nonzero sites after 1 convolution How to keep the same level of sparsity throughout the network? 12.. Table 4: Semantic segmentation performance of five different convolutional networks on the NYU Depth test set (v2) on 40 classes. The table displays the pixel-wise classification accuracy, the computational costs (in FLOPs), and the memory requirements (c.f. Table 2) of each of the models. - "3D Semantic Segmentation with Submanifold Sparse Convolutional Networks"