High-dimensional Convolutional Networks for Geometric Pattern Recognition

Code

• github.com/chrischoy/HighDimConvNets
  OfficialIn paperpytorch★ 40
• github.com/StanfordVL/MinkowskiEngine
  pytorch★ 2,889
• github.com/NVIDIA/MinkowskiEngine
  pytorch★ 2,888
• github.com/shwoo93/minkowskiengine
  pytorch★ 39

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.

Tasks

Reproductions