Gradient-Based Trajectory Optimisation over Continuous Poses for Sparse-View Cone-Beam CT

Key points

• We make every source pose an individual continuous variable and move all poses jointly by gradient ascent on the scanner's kinematic manifold.1
• The objective combines soft-Tuy plane coverage, continuous View Covariance Loss, and an analytic attenuation-aware ray-bundle penalty.1
• Sparseprescan planning benefits from matching prescan and planned acquisition manifolds.1
• Continuous pose optimisation incorporates attenuation and scanner constraints directly into sparse-view acquisition design.1

Sources (1)

• [1]Gradient-Based Trajectory Optimisation over Continuous Poses for Sparse-View Cone-Beam CT
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 07:23 AM

  Trajectory optimisation for cone-beam computed tomography (CT) determines which information sparse-view scans acquire.

  We make every source pose an individual continuous variable and move all poses jointly by gradient ascent on the scanner's kinematic manifold.

Extractive summary: sentences quoted from the sources.