Arnold

Abstract

Arnold is a physically based renderer for feature-length animation and visual effects. Conceived in an era of complex multi-pass rasterization-based workflows struggling to keep up with growing demands for complexity and realism, Arnold was created to take on the challenge of making the simple and elegant approach of brute-force Monte Carlo path tracing practical for production rendering. Achieving this required building a robust piece of ray-tracing software that can ingest large amounts of geometry with detailed shading and lighting and produce images with high fidelity, while scaling well with the available memory and processing power. Arnold’s guiding principles are to expose as few controls as possible, provide rapid feedback to artists, and adapt to various production workflows. In this article, we describe its architecture with a focus on the design and implementation choices made during its evolutionary development to meet the aforementioned requirements and goals. Arnold’s workhorse is a unidirectional path tracer that avoids the use of hard-to-manage and artifact-prone caching and sits on top of a ray-tracing engine optimized to shoot and shade billions of spatially incoherent rays throughout a scene. A comprehensive API provides the means to configure and extend the system’s functionality, to describe a scene, render it, and save the results.

Towards Garment Sewing Pattern Reconstruction from a Single Image

“…For each simulated garment, we render 24 views from cameras uniformly distributed around the human body using Arnold renderer [Georgiev et al 2018]. In some rare cases, the simulator produces inappropriate simulation results, such as with wrong sizes or self-intersection, which are manually removed.…”

An inverse method for the exploration of layered material appearance

“…The case of a stack of rough interfaces and thick media is more complex, even though it remains restricted to the geometric optics regime. Practical implementations in rendering engines (e.g., Arnold renderer [Georgiev et al 2019]) usually take a non-physical approach: each interface is modeled through Microfacet theory and the resulting BRDFs are linearly combined. The most common microfacet distributions are those of Beckmann [1963] and GGX [Walter et al 2007], each relying on a roughness parameter and a ratio of refractive indices.…”