Integrated Forward-Inverse Network for Lensless Image Reconstruction

2026-07-05

AI Summary

Lensless imaging suffers from an ill-posed inverse problem due to the broad point spread function (PSF) and highly aliased measurements, rendering it sensitive to calibration errors and model mismatch. This work proposes a physics-guided multi-scale deep network architecture that alternately embeds differentiable forward projections and learnable inverse updates at each scale, enabling bidirectional coupling between measurement and image domains. This design allows adaptive adjustment of the PSF kernel under model uncertainty. By integrating physical priors, differentiable modeling, and end-to-end training, the method achieves state-of-the-art reconstruction quality across multiple lensless imaging benchmarks, including a newly introduced dataset, and demonstrates strong performance on Gaussian deblurring and simulated in-line holography tasks.

Abstract

Lensless imaging enables compact and versatile computational cameras by replacing bulky optics with thin coded elements. However, reconstruction from the resulting measurements is challenging: large-footprint point-spread functions (PSFs) produce highly multiplexed observations, making inversion severely ill-conditioned and sensitive to calibration errors and model mismatch. While deep learning approaches, including hybrid models that incorporate physics priors, have shown promise, explicitly maintaining data fidelity throughout the network hierarchy remains difficult. Here, we propose the Integrated Forward-Inverse Network (IFIN), a physics-guided architecture that interleaves differentiable forward projections with learnable inverse updates at every scale, enabling complementary cues to be exploited jointly in the measurement and image domains. This bidirectional coupling supports progressive, physics-consistent refinement and permits system-constrained PSF kernel adaptation under model uncertainty. On challenging lensless benchmarks, including a newly introduced dataset, IFIN achieves state-of-the-art reconstruction quality. We further observe competitive performance on Gaussian deblurring and simulated inline holography reconstruction, suggesting that the same interleaving principle can extend beyond lensless cameras.

Problem

• lensless imaging
• image reconstruction
• point-spread function
• ill-conditioned inversion
• model mismatch

Innovation

• lensless imaging
• physics-guided deep learning
• forward-inverse network
• point-spread function adaptation
• computational imaging