Generative World Renderer

TL;DR

This paper introduces a large-scale, dynamic dataset from AAA games for improving generative inverse and forward rendering, enabling better realism, temporal coherence, and controllable video generation in real-world scenarios.

Contribution

The creation of a 4 million frame AAA game dataset with synchronized RGB and G-buffer channels, and a novel VLM-based evaluation protocol for inverse rendering performance.

Findings

Inverse renderers fine-tuned on the dataset show improved cross-dataset generalization.

The VLM-based assessment correlates well with human judgment.

The toolkit allows style editing of AAA games using text prompts.

Abstract

Scaling generative inverse and forward rendering to real-world scenarios is bottlenecked by the limited realism and temporal coherence of existing synthetic datasets. To bridge this persistent domain gap, we introduce a large-scale, dynamic dataset curated from visually complex AAA games. Using a novel dual-screen stitched capture method, we extracted 4M continuous frames (720p/30 FPS) of synchronized RGB and five G-buffer channels across diverse scenes, visual effects, and environments, including adverse weather and motion-blur variants. This dataset uniquely advances bidirectional rendering: enabling robust in-the-wild geometry and material decomposition, and facilitating high-fidelity G-buffer-guided video generation. Furthermore, to evaluate the real-world performance of inverse rendering without ground truth, we propose a novel VLM-based assessment protocol measuring semantic, spatial, and temporal consistency. Experiments demonstrate that inverse renderers fine-tuned on our data achieve superior cross-dataset generalization and controllable generation, while our VLM evaluation strongly correlates with human judgment. Combined with our toolkit, our forward renderer enables users to edit styles of AAA games from G-buffers using text prompts.