Dimensions overall score 7.0

GitHub Code Pulse

No public code linked for this paper yet.

• The paper introduces a large-scale, dynamic dataset of 4M continuous frames (720p/30 FPS) with synchronized RGB and five G-buffer channels, curated from AAA games.method 95%
• The dataset includes diverse scenes, visual effects, and environments, including adverse weather and motion-blur variants.technical 90%
• The dataset enables robust in-the-wild geometry and material decomposition for bidirectional rendering.result 85%
• The dataset facilitates high-fidelity G-buffer-guided video generation.result 85%
• The paper proposes a novel VLM-based assessment protocol for evaluating inverse rendering performance without ground truth.method 90%
• Inverse renderers fine-tuned on their data achieve superior cross-dataset generalization and controllable generation.result 80%
• The VLM evaluation protocol strongly correlates with human judgment.result 80%
• The forward renderer enables users to edit styles of AAA games from G-buffers using text prompts.technical 85%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card