Unifying Deep Stochastic Processes for Image Enhancement

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.

AI is excellent at image enhancement—turning blurry or noisy photos into clear images using complex math. However, scientists have created so many different versions of this process that the field has become chaotic, making it hard to know which method is truly best. To solve this, we analyzed the underlying math and discovered a unified theory: despite looking different, almost all these AI tools share the exact same foundation. This allowed us to group the methods into three families and test them on a completely level playing field. Our tests revealed that no single method is always best. Instead, performance depends on subtle, specific mathematical tweaks. To help developers compare these methods fairly, we released a free, open-source software library called ItoVision. This tool allows anyone to easily test different mathematical choices, helping the community build better, faster photo-fixing AI.