Image Labeling by Assignment
arXiv:1603.05285 · doi:10.1007/s10851-016-0702-4
Abstract
We introduce a novel geometric approach to the image labeling problem. Abstracting from specific labeling applications, a general objective function is defined on a manifold of stochastic matrices, whose elements assign prior data that are given in any metric space, to observed image measurements. The corresponding Riemannian gradient flow entails a set of replicator equations, one for each data point, that are spatially coupled by geometric averaging on the manifold. Starting from uniform assignments at the barycenter as natural initialization, the flow terminates at some global maximum, each of which corresponds to an image labeling that uniquely assigns the prior data. Our geometric variational approach constitutes a smooth non-convex inner approximation of the general image labeling problem, implemented with sparse interior-point numerics in terms of parallel multiplicative updates that converge efficiently.
References in corpus (2)
Cited by in corpus (14)
- Geometric Numerical Integration of the Assignment Flow
- Iterative Multiplicative Filters for Data Labeling
- A Graph Framework for Manifold-valued Data
- Assignment Flows for Data Labeling on Graphs: Convergence and Stability
- Image Labeling Based on Graphical Models Using Wasserstein Messages and Geometric Assignment
- Unsupervised Assignment Flow: Label Learning on Feature Manifolds by Spatially Regularized Geometric Assignment
- Textually Enriched Neural Module Networks for Visual Question Answering
- Measure-Valued Variational Models with Applications to Diffusion-Weighted Imaging
- Assignment Flow for Order-Constrained OCT Segmentation
- A Transportation Distance for Signal Analysis
- On the Geometric Mechanics of Assignment Flows for Metric Data Labeling
- A Nonlocal Graph-PDE and Higher-Order Geometric Integration for Image Labeling
- Quantum State Assignment Flows
- Elements of Convex Geometry in Hadamard Manifolds with Application to Equilibrium Problems