AI RESEARCH
On the Degrees of Freedom of Gridded Control Points in Learning-Based Medical Image Registration
arXiv CS.CV
•
ArXi:2603.16940v1 Announce Type: cross Many registration problems are ill-posed in homogeneous or noisy regions, and dense voxel-wise decoders can be unnecessarily high-dimensional. A sparse control-point parameterisation provides a compact, smooth deformation representation while reducing memory and improving stability. This work investigates the required control points for learning-based registration network development. We present GridReg, a learning-based registration framework that replaces dense voxel-wise decoding with displacement predictions at a sparse grid of control points.