DefTransNet, a Transformer-based point cloud registration architecture with a learnable transformation matrix and graph features, reports lower mean distance errors than baselines across four synthetic and real-world datasets.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DefTransNet: A Transformer-based Method for Non-Rigid Point Cloud Registration in the Simulation of Soft Tissue Deformation
DefTransNet, a Transformer-based point cloud registration architecture with a learnable transformation matrix and graph features, reports lower mean distance errors than baselines across four synthetic and real-world datasets.