Unifying Deep Stochastic Processes for Image Enhancement
For researchers in image enhancement, this provides a unified perspective and a modular library (ItoVision) to facilitate fair comparison and prototyping, though the findings are largely negative (no single best method).
The paper unifies various stochastic processes for image enhancement into a common SDE framework, showing that methods differ mainly in drift, diffusion, terminal distributions, and boundary conditions. A controlled empirical study found no consistently dominant method, but identified key design choices affecting performance.
Deep stochastic processes have recently become a central paradigm for image enhancement, with many methods explicitly conditioning the stochastic trajectory on the degraded input. However, the relationship between these conditional processes and standard diffusion models remains unclear. In this work, we introduce a unified perspective on stochastic image enhancement by classifying recent methods into three families of continuous-time processes: unconditional diffusion models, Ornstein-Uhlenbeck (OU) processes, and diffusion bridges. We show that all of these approaches arise from a common stochastic differential equation (SDE) formulation. This framework makes explicit that seemingly disparate methods differ primarily in their drift and diffusion terms, terminal distributions, and boundary conditions, while schedulers and samplers constitute orthogonal design choices. Leveraging this unification, we conduct a controlled empirical study across multiple image enhancement tasks using identical architectures and training protocols. Our results reveal no consistently dominant method; instead, we identify and disentangle the specific design choices that most strongly influence performance. Finally, we release ItoVision, a modular PyTorch library that implements the unified framework and enables rapid prototyping and fair comparison of stochastic image enhancement methods.