Skip to main content
. Author manuscript; available in PMC: 2023 Jul 10.
Published in final edited form as: Int J Comput Assist Radiol Surg. 2023 Apr 20;18(6):1025–1032. doi: 10.1007/s11548-023-02893-3

Fig. 2.

Fig. 2

LiverMatch network overview: 1) Encoder - down-samples input point clouds and extract associated features; 2) Transformer - updates features to conditioned features with self-global geometry and cross-global geometry information. 3) Decoder - up-samples conditioned features to obtain per-point features. 4) Matching - calculates a confidence matrix to select matches. An additional 1D convolution decodes the xS to visibility scores.