BAM! Bayesian Anything Model: a foundation model for generative computational imaging

2026-09-30

AI Summary

This study addresses the absence of physics-aware foundation models in computational imaging, where existing methods suffer from significant bias and poor generalization. We propose BAM, a lightweight foundation model built upon an operator-conditioned RAM backbone. By jointly pretraining on large-scale image corpora and a forward operator library, BAM achieves few-step posterior sampling via conditional flow matching, enabling direct specification of instrument physics at inference. With only 36M parameters, the model eliminates the need for likelihood approximation or guidance weight tuning, supporting zero-shot generalization to unseen tasks and data. Experimental results demonstrate that merely three sampling steps suffice to outperform both task-specific and zero-shot state-of-the-art methods, substantially reducing computational costs while improving sample quality.

Abstract

Generative models are transforming Bayesian computational imaging, yet the field still lacks physics-aware foundation models. Current practice falls into two camps. Large foundation image models are deployed as plug-and-play priors with zero-shot approximate likelihood guidance, which introduces significant bias and computational cost. Physics-aware generative models avoid this bias, but each is tied to a specific dataset, task and instrument. We introduce BAM (Bayesian Anything Model), a lightweight foundation model for few-step, physics-aware posterior sampling that generalises robustly to unseen data and tasks, zero-shot or with minimal finetuning. BAM upgrades the operator-conditioned Reconstruct Anything Model (RAM) backbone (Terris et al.) into a conditional flow map, so instrument physics is specified at inference time rather than fixed during training. BAM has just 36M parameters and is pre-trained jointly on large image corpora and libraries of forward operators. A single network then draws posterior samples in a few steps, with no likelihood approximation and no guidance weights to tune. Across linear inverse problems on FFHQ, AFHQ, LSUN, DIV2K and the Kohler camera-shake benchmark, BAM outperforms in just 3 steps both specialised models and leading zero-shot methods in sample quality, at a fraction of their computational cost. BAM gives the community an accessible entry point to generative computational imaging, lowers the economic and environmental cost of training imaging models, and opens a new path for research on physics-aware Bayesian computational imaging. Official page: https://bayesian-anything-model.github.io/

Problem

• Bayesian computational imaging
• foundation model
• physics-aware generative models
• inverse problems
• posterior sampling

Innovation

• Bayesian computational imaging
• foundation model
• conditional flow map
• posterior sampling
• physics-aware