When Poisson-count data fall outside the image cone, likelihood maximizers and ML-EM cluster points concentrate on sparse point masses, while inside the cone they retain full support.
Introduction to inverse problems in imaging
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The ML-EM algorithm in continuum: sparse measure solutions
When Poisson-count data fall outside the image cone, likelihood maximizers and ML-EM cluster points concentrate on sparse point masses, while inside the cone they retain full support.