New coupled optimal transport framework uses landmarks to guide distribution transformations

Researchers have introduced a new coupled optimal transport (OT) framework that uses a small set of annotated landmarks to guide the identification of transformations between data distributions. This approach integrates the optimization of the transport plan and the deformation field into a single model, ensuring the deformation is influenced by both landmark data and cost-driven distribution matching. The framework establishes a connection between landmark-based registration and transport-based distribution matching, allowing for the recovery of transport maps with sparse geometric supervision. A numerical algorithm has been developed for computation, and its effectiveness has been demonstrated in shape matching applications. AI

IMPACT This research could improve shape matching and geometric transformation recovery in machine learning applications.

RANK_REASON The cluster contains a research paper detailing a new computational framework. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• Hugging Face
• Landmark Constraints
• optimal transport

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →