Noise-Aware Merging of High Dynamic Range Image Stacks without Camera Calibration

TL;DR

This paper presents a noise-aware method for merging HDR image stacks that does not require camera calibration, using a Poisson noise estimator for unbiased radiance estimation across various cameras.

Contribution

It introduces a calibration-free Poisson noise estimator for HDR merging, simplifying the process while maintaining accuracy across different camera types.

Findings

Unbiased radiance estimation with a simple Poisson noise model.

Consistent results across simulated and real images.

Effective for diverse camera types from smartphones to professional cameras.

Abstract

A near-optimal reconstruction of the radiance of a High Dynamic Range scene from an exposure stack can be obtained by modeling the camera noise distribution. The latent radiance is then estimated using Maximum Likelihood Estimation. But this requires a well-calibrated noise model of the camera, which is difficult to obtain in practice. We show that an unbiased estimation of comparable variance can be obtained with a simpler Poisson noise estimator, which does not require the knowledge of camera-specific noise parameters. We demonstrate this empirically for four different cameras, ranging from a smartphone camera to a full-frame mirrorless camera. Our experimental results are consistent for simulated as well as real images, and across different camera settings.