研究目的
To develop an unbiased and flexible layered BSDF model that accurately simulates light transport in layered materials with arbitrary interfaces and internal media, overcoming limitations of previous analytical or discretized approaches.
研究成果
The introduced layered BSDF model provides an unbiased and flexible solution for simulating light transport in layered materials, supporting arbitrary interfaces, volumetric scattering, anisotropy, and spatial variation. It outperforms previous methods in accuracy and editability, though it has computational overhead and limitations in handling global geometric effects. Future work could extend to BSSRDFs and integrate wave optics.
研究不足
The model assumes thin flat layers and negligible horizontal light spreading, so it cannot capture global-scale effects like internal caustics, shadowing from major normal variations, or color bleeding in varying media. Performance degrades with optically thick layers and many scattering events, and wave optics effects (e.g., thin film interference) are not handled.
1：Experimental Design and Method Selection:
The methodology involves a position-free path integral formulation for light transport in flat slabs, using Monte Carlo simulation techniques including unidirectional path tracing with next-event estimation and a bidirectional estimator. This avoids the high-variance geometry terms in standard formulations.
2：Sample Selection and Data Sources:
The experiments use synthetic layered material configurations with varying interface properties (e.g., microfacet BSDFs) and volumetric scattering media (e.g., anisotropic phase functions), without specific real-world datasets.
3：List of Experimental Equipment and Materials:
The primary tool is a computer with a CPU (e.g., Intel i7-6800K) running the Mitsuba renderer for simulation and rendering. No physical equipment is used as it is a computational study.
4：Experimental Procedures and Operational Workflow:
For each BSDF query, Monte Carlo paths are sampled by simulating light interactions within layers, evaluating contributions, and averaging samples. The process includes importance sampling, pdf evaluation, and multiple importance sampling for variance reduction.
5：Data Analysis Methods:
Results are validated through cross-validation with ground truth (e.g., binning methods), white furnace tests for energy conservation, and performance comparisons with baseline models. Statistical analysis involves measuring render times and visual quality assessments.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
