Self-Calibrating, Fully Differentiable NLOS Inverse Rendering
This addresses reconstruction ambiguities and noise in NLOS imaging, which is incremental as it builds on existing methods by adding self-calibration and differentiability.
The paper tackles the problem of reconstruction artifacts in time-resolved non-line-of-sight (NLOS) imaging by introducing a fully-differentiable end-to-end pipeline that self-calibrates imaging parameters during reconstruction, resulting in robust geometry and albedo recovery even under significant noise.
Existing time-resolved non-line-of-sight (NLOS) imaging methods reconstruct hidden scenes by inverting the optical paths of indirect illumination measured at visible relay surfaces. These methods are prone to reconstruction artifacts due to inversion ambiguities and capture noise, which are typically mitigated through the manual selection of filtering functions and parameters. We introduce a fully-differentiable end-to-end NLOS inverse rendering pipeline that self-calibrates the imaging parameters during the reconstruction of hidden scenes, using as input only the measured illumination while working both in the time and frequency domains. Our pipeline extracts a geometric representation of the hidden scene from NLOS volumetric intensities and estimates the time-resolved illumination at the relay wall produced by such geometric information using differentiable transient rendering. We then use gradient descent to optimize imaging parameters by minimizing the error between our simulated time-resolved illumination and the measured illumination. Our end-to-end differentiable pipeline couples diffraction-based volumetric NLOS reconstruction with path-space light transport and a simple ray marching technique to extract detailed, dense sets of surface points and normals of hidden scenes. We demonstrate the robustness of our method to consistently reconstruct geometry and albedo, even under significant noise levels.