Research map / Learning theory and algorithms / Optimal transport

Variational Wasserstein Model on Riemannian Manifolds for Image Segmentation

NeurIPS 2026 · Learning theory and algorithms › Optimal transport · approach: Transport map estimation

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes an OT-based variational image segmentation model using Wasserstein distances on Riemannian manifolds to improve fidelity when foreground and background appearances are similar.

Problem

Variational image segmentation relies on data fidelity terms; existing optimal transport (OT)-based models match feature distributions using color or intensity similarity, which fails when foreground and background share similar appearance.

Approach

Extends OT-based variational segmentation by formulating the Wasserstein fidelity on Riemannian manifolds so that matching uses richer geometric structure beyond raw color or intensity. Only the abstract was available, so details are limited.

Closest papers by meaning

• Integrating Efficient Optimal Transport and Functional Maps For Unsupervised Shape Correspondence Learning (CVPR 2024 · similarity 0.87)
• Riemannian Neural Optimal Transport (ICML 2026 · similarity 0.87)
• Geometry-Aware Optimal Transport: Fast Intrinsic Dimension and Wasserstein Distance Estimation (NeurIPS 2026 · similarity 0.85)
• Linear Spherical Sliced Optimal Transport: A Fast Metric for Comparing Spherical Data (ICLR 2025 · similarity 0.85)
• Wasserstein Wormhole: Scalable Optimal Transport Distance with Transformer (ICML 2024 · similarity 0.84)
• Task-Induced Riemannian Metrics for Vision Transformer Feature Spaces (NeurIPS 2026 · similarity 0.84)