Event Ellipsometer: Event-based Mueller-Matrix Video Imaging

TL;DR

The paper addresses the challenge of capturing dynamic polarization information by proposing Event Ellipsometer, a system that uses an event camera and fast-rotating quarter-wave plates to acquire Mueller-matrix videos at 30 fps. It introduces an ellipsometric-event image formation model, a calibration protocol, and a robust two-stage reconstruction (per-pixel estimation plus spatio-temporal propagation) to recover per-pixel Mueller matrices under HDR and motion. The approach is validated on synthetic and real data, demonstrating accurate Mueller matrices in dynamic scenes and enabling applications in photoelasticity, transparent-material detection, and HDR polarization imaging. This work extends ellipsometry to dynamic environments, enabling qualitative and quantitative polarization analysis in previously intractable settings with high temporal resolution.

Abstract

Light-matter interactions modify both the intensity and polarization state of light. Changes in polarization, represented by a Mueller matrix, encode detailed scene information. Existing optical ellipsometers capture Mueller-matrix images; however, they are often limited to capturing static scenes due to long acquisition times. Here, we introduce Event Ellipsometer, a method for acquiring a Mueller-matrix video for dynamic scenes. Our imaging system employs fast-rotating quarter-wave plates (QWPs) in front of a light source and an event camera that asynchronously captures intensity changes induced by the rotating QWPs. We develop an ellipsometric-event image formation model, a calibration method, and an ellipsometric-event reconstruction method. We experimentally demonstrate that Event Ellipsometer enables Mueller-matrix video imaging at 30fps, extending ellipsometry to dynamic scenes.

Figures (9)

• Figure 1: Overview of Event Ellipsometer. (a) Our imaging system captures the Mueller matrix at 30 fps from event streams induced by the continuously rotating QWPs. (b) With our experimental prototype, (c) we demonstrate ellipsometric analysis for dynamic scenes and various applications.
• Figure 2: Imaging system of Event Ellipsometer. (a) Schematic diagram illustrating the optical arrangement and hardware operation. (b) Timeline showing the rotation of two QWPs and the event measurement. (c) Our hardware prototype. The system can move the light source and camera position for use in (d) Reflection mode or (e) Transmission mode.
• Figure 3: Overview of our Mueller-matrix reconstruction pipeline. This method consists of two steps: (1) per-pixel reconstruction and (2) propagation and refinement.
• Figure 4: Synthetic data evaluation result. (a) The rendered images include two materials: blue silicone and brass. (b) The plot shows the error (mean absolute error) of the reconstructed Mueller matrix over the number of iterations. (c) Ground truth Mueller matrix. (d)&(e) Top: the reconstructed Mueller-matrix images for the SVD initialization and the full-stage, respectively. Pixels with insufficient event counts are visualized in white. Bottom: a plot of the differentiation of log intensity and fitted line with our method.
• Figure 5: Assessment of reconstructed Mueller matrix on real data. (a) Evaluation with known optical elements. We show the corresponding mean squared errors (MSEs). (b) Measurement on an in-the-wild metal plate induces strong diagonal components.