ePBR: Extended PBR Materials in Image Synthesis

TL;DR

The paper addresses the gap between physically based realism and efficient image synthesis by extending intrinsic image representations to include both reflection and transmission through ePBR materials. It introduces an explicit intrinsic compositing framework and a full ePBR material BSDF that blends diffuse, specular, and specular transmission terms under a thin-surface assumption. The intrinsic channels capture geometry, albedo, roughness, metallicity, transparency, and screen-space illumination, enabling deterministic, editable rendering without full path tracing. The approach yields faithful rendering of high specular regions and offers memory-friendly, real-time capabilities, though limitations include anisotropy, subsurface scattering, and more complex multi-reflection effects.

Abstract

Realistic indoor or outdoor image synthesis is a core challenge in computer vision and graphics. The learning-based approach is easy to use but lacks physical consistency, while traditional Physically Based Rendering (PBR) offers high realism but is computationally expensive. Intrinsic image representation offers a well-balanced trade-off, decomposing images into fundamental components (intrinsic channels) such as geometry, materials, and illumination for controllable synthesis. However, existing PBR materials struggle with complex surface models, particularly high-specular and transparent surfaces. In this work, we extend intrinsic image representations to incorporate both reflection and transmission properties, enabling the synthesis of transparent materials such as glass and windows. We propose an explicit intrinsic compositing framework that provides deterministic, interpretable image synthesis. With the Extended PBR (ePBR) Materials, we can effectively edit the materials with precise controls.

Figures (7)

• Figure 1: Highly specular and transparent objects are very common in the real world.
• Figure 2: Thin surface assumption: Ignoring the internal reflection, light traveling through a transparent thin surface refracts twice as it enters and exits, and reflects once only on the top surface. For a smooth surface, light exits with the same direction as it enters and the offset could be ignored.
• Figure 3: Three typical materials: Metal ($t=0, m=1$), Dielectric ($t=0, m=0$) and Glass ($t=1, m=0$).
• Figure 4: Roughness evaluation. Ours: Directly apply filtering kernel to the specular reflection image; GT: Path tracing with Monte Carlo sampling.
• Figure 5: Generate $\bm{\mathrm{A}}_\mathrm{mr}$ from SSRT. To get a mirror-like reflection image for an interest region, using SSRT to trace the ray and find the corresponding color.