# A Generalizable Light Transport 3D Embedding for Global Illumination

arXiv: https://arxiv.org/abs/2510.18189 · HTML: https://mlsanity.com/paper/2510.18189/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 55 references that are in the map.
- First posted on arXiv 21 Oct 2025.

## Builds on (most cited first, 20 of 20)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [Instant Neural Graphics Primitives with a Multiresolution Hash Encoding](https://mlsanity.com/paper/2201.05989/) (2022)
- [Dynamic Graph CNN for Learning on Point Clouds](https://mlsanity.com/paper/1801.07829/) (2018)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [PointConv: Deep Convolutional Networks on 3D Point Clouds](https://mlsanity.com/paper/1811.07246/) (2018)
- [Point Transformer V3: Simpler, Faster, Stronger](https://mlsanity.com/paper/2312.10035/) (2023)
- [Infinigen Indoors: Photorealistic Indoor Scenes using Procedural Generation](https://mlsanity.com/paper/2406.11824/) (2024)
- [Real-time Neural Radiance Caching for Path Tracing](https://mlsanity.com/paper/2106.12372/) (2021)
- [Meta 3D AssetGen: Text-to-Mesh Generation with High-Quality Geometry, Texture, and PBR Materials](https://mlsanity.com/paper/2407.02445/) (2024)
- [Monte Carlo Convolution for Learning on Non-Uniformly Sampled Point Clouds](https://mlsanity.com/paper/1806.01759/) (2018)
- [Deep Shading: Convolutional Neural Networks for Screen-Space Shading](https://mlsanity.com/paper/1603.06078/) (2016)
- [Crowdsourcing in Computer Vision](https://mlsanity.com/paper/1611.02145/) (2016)
- [RenderFormer: Transformer-based Neural Rendering of Triangle Meshes with Global Illumination](https://mlsanity.com/paper/2505.21925/) (2025)
- [Active Exploration for Neural Global Illumination of Variable Scenes](https://mlsanity.com/paper/2203.08272/) (2022)
- [Neural Free-Viewpoint Relighting for Glossy Indirect Illumination](https://mlsanity.com/paper/2307.06335/) (2023)
- [Deep-learning the Latent Space of Light Transport](https://mlsanity.com/paper/1811.04756/) (2018)
- [Learning Neural Light Transport](https://mlsanity.com/paper/2006.03427/) (2020)
- [GPU-friendly, Parallel, and (Almost-)In-Place Construction of Left-Balanced k-d Trees](https://mlsanity.com/paper/2211.00120/) (2022)

## Built on by (most cited first, 2 of 2)

- [Point2Radio: A Foundation Model for Cross-Scene Radio Fields from Material-Aware Point Clouds](https://mlsanity.com/paper/2607.28994/) (2026)
- [RenderFormer-V2: Neural Rendering with Heterogeneous Scene Primitives](https://mlsanity.com/paper/2609.05738/) (2026)

## Abstract (arXiv)

Global illumination (GI) is essential for realistic rendering but remains computationally expensive due to the complexity of simulating indirect light transport. Recent neural methods have mainly relied on per-scene optimization, sometimes extended to handle changes in camera or geometry. Efforts toward cross-scene generalization have largely stayed in 2D screen space, such as neural denoising or G-buffer based GI prediction, which often suffer from view inconsistency and limited spatial understanding. We propose a generalizable 3D light transport embedding that approximates global illumination directly from 3D scene configurations, without using rasterized or path-traced cues. Each scene is represented as a point cloud with geometric and material features. A scalable transformer models global point-to-point interactions to encode these features into neural primitives. At render time, each query point retrieves nearby primitives via nearest-neighbor search and aggregates their latent features through cross-attention to predict the desired rendering quantity. We demonstrate results on diffuse global illumination prediction across diverse indoor scenes with varying layouts, geometry, and materials. The embedding trained for irradiance estimation can be quickly adapted to new rendering tasks with limited fine-tuning. We also present preliminary results for spatial-directional radiance field estimation for glossy materials and show how the normalized field can accelerate unbiased path guiding. This approach highlights a path toward integrating learned priors into rendering pipelines without explicit ray-traced illumination cues.