High-dimensional Convolutional Networks for Geometric Pattern Recognition

• URL: http://arxiv.org/abs/2005.08144v1
• Date: Sun, 17 May 2020 01:46:12 GMT
• Title: High-dimensional Convolutional Networks for Geometric Pattern Recognition
• Authors: Christopher Choy, Junha Lee, Rene Ranftl, Jaesik Park, Vladlen Koltun
• Abstract summary: We present high-dimensional convolutional networks (ConvNets) for pattern recognition problems. We first study the effectiveness of convolutional networks in detecting linear subspaces in high-dimensional spaces with up to 32 dimensions. We then apply high-dimensional ConvNets to 3D registration under rigid motions and image correspondence estimation.
• Score: 75.43345656210992
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Many problems in science and engineering can be formulated in terms of geometric patterns in high-dimensional spaces. We present high-dimensional convolutional networks (ConvNets) for pattern recognition problems that arise in the context of geometric registration. We first study the effectiveness of convolutional networks in detecting linear subspaces in high-dimensional spaces with up to 32 dimensions: much higher dimensionality than prior applications of ConvNets. We then apply high-dimensional ConvNets to 3D registration under rigid motions and image correspondence estimation. Experiments indicate that our high-dimensional ConvNets outperform prior approaches that relied on deep networks based on global pooling operators.

Related papers

• How Low Can You Go? Searching for the Intrinsic Dimensionality of Complex Networks using Metric Node Embeddings [1.7061868168035932]
  Low-dimensional embeddings are essential for machine learning tasks involving graphs. We prove that lower dimensional embeddings are possible when using Euclidean metric embeddings. For the first time, we demonstrate that even large-scale networks can be effectively embedded in very low-dimensional spaces.
  arXiv  Detail & Related papers  (2025-03-03T16:37:38Z)
• Geometry Distributions [51.4061133324376]
  We propose a novel geometric data representation that models geometry as distributions. Our approach uses diffusion models with a novel network architecture to learn surface point distributions. We evaluate our representation qualitatively and quantitatively across various object types, demonstrating its effectiveness in achieving high geometric fidelity.
  arXiv  Detail & Related papers  (2024-11-25T04:06:48Z)
• GraphCSPN: Geometry-Aware Depth Completion via Dynamic GCNs [49.55919802779889]
  We propose a Graph Convolution based Spatial Propagation Network (GraphCSPN) as a general approach for depth completion. In this work, we leverage convolution neural networks as well as graph neural networks in a complementary way for geometric representation learning. Our method achieves the state-of-the-art performance, especially when compared in the case of using only a few propagation steps.
  arXiv  Detail & Related papers  (2022-10-19T17:56:03Z)
• Deep Learning Assisted Optimization for 3D Reconstruction from Single 2D Line Drawings [13.532686360047574]
  We propose to train deep neural networks to detect pairwise relationships among geometric entities in 3D objects. Experiments on a large dataset of CAD models show that, by leveraging deep learning in a geometric constraint solving pipeline, the success rate of optimization-based 3D reconstruction can be significantly improved.
  arXiv  Detail & Related papers  (2022-09-06T17:59:11Z)
• Laplacian2Mesh: Laplacian-Based Mesh Understanding [4.808061174740482]
  We introduce a novel and flexible convolutional neural network (CNN) model, called Laplacian2Mesh, for 3D triangle mesh. Mesh pooling is applied to expand the receptive field of the network by the multi-space transformation of Laplacian. Experiments on various learning tasks applied to 3D meshes demonstrate the effectiveness and efficiency of Laplacian2Mesh.
  arXiv  Detail & Related papers  (2022-02-01T10:10:13Z)
• Putting 3D Spatially Sparse Networks on a Diet [21.881294733075393]
  We propose a compact weight-sparse and spatially sparse 3D convnet (WS3-ConvNet) for semantic segmentation and instance segmentation. We employ various network pruning strategies to find compact networks and show our WS3-ConvNet achieves minimal loss in performance (2-15% drop) with orders-of-15% smaller number of parameters (1/100 compression rate) Finally, we systematically analyze the compression patterns of WS3-ConvNet and show interesting emerging sparsity patterns common in our compressed networks to further speed up inference.
  arXiv  Detail & Related papers  (2021-12-02T15:20:15Z)
• UnProjection: Leveraging Inverse-Projections for Visual Analytics of High-Dimensional Data [63.74032987144699]
  We present NNInv, a deep learning technique with the ability to approximate the inverse of any projection or mapping. NNInv learns to reconstruct high-dimensional data from any arbitrary point on a 2D projection space, giving users the ability to interact with the learned high-dimensional representation in a visual analytics system.
  arXiv  Detail & Related papers  (2021-11-02T17:11:57Z)
• Deep Active Surface Models [60.027353171412216]
  Active Surface Models have a long history of being useful to model complex 3D surfaces but only Active Contours have been used in conjunction with deep networks. We introduce layers that implement them that can be integrated seamlessly into Graph Convolutional Networks to enforce sophisticated smoothness priors.
  arXiv  Detail & Related papers  (2020-11-17T18:48:28Z)
• Graph Signal Processing for Geometric Data and Beyond: Theory and Applications [55.81966207837108]
  Graph Signal Processing (GSP) enables processing signals that reside on irregular domains. GSP methodologies for geometric data in a unified manner by bridging the connections between geometric data and graphs. Recently developed Graph Neural Networks (GNNs) interpret the operation of these networks from the perspective of GSP.
  arXiv  Detail & Related papers  (2020-08-05T03:20:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.