Graph Canvas for Controllable 3D Scene Generation
This addresses the need for more controllable and adaptable 3D scene generation for AI systems interacting with the physical world, though it appears incremental as it builds on existing graph-based and in-context learning methods.
The paper tackles the problem of limited adaptability in 3D scene generation by introducing GraphCanvas3D, a framework that uses hierarchical graph descriptions and in-context learning to enable dynamic scene manipulation without retraining, and it demonstrates enhanced usability, flexibility, and adaptability in experiments.
Spatial intelligence is foundational to AI systems that interact with the physical world, particularly in 3D scene generation and spatial comprehension. Current methodologies for 3D scene generation often rely heavily on predefined datasets, and struggle to adapt dynamically to changing spatial relationships. In this paper, we introduce GraphCanvas3D, a programmable, extensible, and adaptable framework for controllable 3D scene generation. Leveraging in-context learning, GraphCanvas3D enables dynamic adaptability without the need for retraining, supporting flexible and customizable scene creation. Our framework employs hierarchical, graph-driven scene descriptions, representing spatial elements as graph nodes and establishing coherent relationships among objects in 3D environments. Unlike conventional approaches, which are constrained in adaptability and often require predefined input masks or retraining for modifications, GraphCanvas3D allows for seamless object manipulation and scene adjustments on the fly. Additionally, GraphCanvas3D supports 4D scene generation, incorporating temporal dynamics to model changes over time. Experimental results and user studies demonstrate that GraphCanvas3D enhances usability, flexibility, and adaptability for scene generation. Our code and models are available on the project website: https://github.com/ILGLJ/Graph-Canvas.