RenderFormer-V2: Neural Rendering with Heterogeneous Scene Primitives
RenderFormer-V2 is a transformer-based neural renderer handling caustics, volumetric scattering and out-of-distribution materials without per-scene training or specialized code.
RenderFormer-V2 is a learned transformer-based neural rendering model that models global light transport as a sequence-to-sequence transformation, handling caustics, volumetric scattering, environment lighting, textured and displaced surfaces and out-of-distribution materials. It uses a two-stage process: a view-independent stage resolving primitive-to-primitive transport, and a view-dependent stage converting the neural scene representation into pixels. Improvements include combined windowed-attention with a rendering-informed attention sink for scalability, support for heterogeneous primitives like environment maps and participating media, and a surface-reflectance-independent neural material encoding, validated across diverse scenes with extensive ablations.
- Models global light transport as sequence-to-sequence transformation
- Two-stage design separates view-independent transport from view-dependent pixelization
- Windowed attention plus rendering-informed attention sink improves scalability
- Supports environment maps, participating media and BRDF-agnostic material encoding
- AI models
- RenderFormer-V2
From the original article
1 min read at huggingface.co
We present 'RenderFormer-V2', a unified learned transformer-based neural rendering model, complementary to modern physics-based rendering systems, that can handle diverse light-transport effects such as caustics, volumetric scattering, environment lighting, textured and displaced surfaces and out-of-distribution materials without per-scene training or specialized code.
By Chong Zeng, Yue Dong, Pieter Peers · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.