<p>Existing non-rigid point cloud registration algorithms often encounter matching errors arise when dealing with similar regions in point clouds, and it is difficult to promptly filter out these misaligned corresponding points. To address these issues, a non-rigid point cloud registration algorithm based on position-aware feature matching is proposed. Firstly, the relative positions of the point cloud are encoded using Fourier transform, decomposing the 3D point cloud of non-rigid objects into feature space information and 3D positional information. This enables non-rigid point cloud registration for similar parts while preventing the loss of positional information during network iterations. Secondly, global information is summarized through the self-attention layers of the transformation blocks, and point cloud information exchange is facilitated through cross-attention layers to promote feature matching between the source and target point clouds. Next, we design and integrate an outlier removal strategy into a high-dimensional convolutional neural network to eliminate incorrect matching correspondences. The Welsch function is applied in the regularization term of the loss function to enhance the algorithm’s robustness against noise and partially overlapping point clouds. Finally, comparative experiments with seven existing algorithms on the 4Dmatch/4Dlomatch dataset demonstrate that our proposed method outperforms the second-best algorithm by 3.3<InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="530_2024_1657_Article_IEq1.gif" Format="GIF" Height="13" Rendition="HTML" Resolution="72" Type="Linedraw" Width="19" /> </InlineMediaObject> <EquationSource Format="TEX">\(-\)</EquationSource> <EquationSource Format="MATHML"><math> <mo>-</mo> </math></EquationSource> </InlineEquation>6.3% in the correspondence index (IR) and 19.15<InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="530_2024_1657_Article_IEq2.gif" Format="GIF" Height="13" Rendition="HTML" Resolution="72" Type="Linedraw" Width="19" /> </InlineMediaObject> <EquationSource Format="TEX">\(-\)</EquationSource> <EquationSource Format="MATHML"><math> <mo>-</mo> </math></EquationSource> </InlineEquation>21.57% in the registration result index (Accr). The experimental results indicate that our method can effectively handle feature-similar regions in point clouds, promptly filter out misaligned corresponding points, and produce more accurate registration results, especially for lower overlap rates.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Position-aware feature matching algorithm based non-rigid point cloud registration

  • Ronqi Wang,
  • Ronguo Zhang,
  • Jing Hu,
  • Rui Zhang,
  • Lifang Wang,
  • Xiaojun Liu

摘要

Existing non-rigid point cloud registration algorithms often encounter matching errors arise when dealing with similar regions in point clouds, and it is difficult to promptly filter out these misaligned corresponding points. To address these issues, a non-rigid point cloud registration algorithm based on position-aware feature matching is proposed. Firstly, the relative positions of the point cloud are encoded using Fourier transform, decomposing the 3D point cloud of non-rigid objects into feature space information and 3D positional information. This enables non-rigid point cloud registration for similar parts while preventing the loss of positional information during network iterations. Secondly, global information is summarized through the self-attention layers of the transformation blocks, and point cloud information exchange is facilitated through cross-attention layers to promote feature matching between the source and target point clouds. Next, we design and integrate an outlier removal strategy into a high-dimensional convolutional neural network to eliminate incorrect matching correspondences. The Welsch function is applied in the regularization term of the loss function to enhance the algorithm’s robustness against noise and partially overlapping point clouds. Finally, comparative experiments with seven existing algorithms on the 4Dmatch/4Dlomatch dataset demonstrate that our proposed method outperforms the second-best algorithm by 3.3 \(-\) - 6.3% in the correspondence index (IR) and 19.15 \(-\) - 21.57% in the registration result index (Accr). The experimental results indicate that our method can effectively handle feature-similar regions in point clouds, promptly filter out misaligned corresponding points, and produce more accurate registration results, especially for lower overlap rates.