Dreamland: Controllable World Creation with Simulator and Generative Models

TL;DR

Dreamland introduces a hybrid framework combining physics-based simulation and large-scale generative models to enable controllable, realistic world creation, improving scene editing and embodied AI training.

Contribution

It proposes a layered world abstraction bridging simulators and generative models, enhancing controllability and realism in synthetic world generation.

Findings

50.8% improved image quality over baselines

17.9% stronger controllability

Effective for training embodied AI agents

Abstract

Large-scale video generative models can synthesize diverse and realistic visual content for dynamic world creation, but they often lack element-wise controllability, hindering their use in editing scenes and training embodied AI agents. We propose Dreamland, a hybrid world generation framework combining the granular control of a physics-based simulator and the photorealistic content output of large-scale pretrained generative models. In particular, we design a layered world abstraction that encodes both pixel-level and object-level semantics and geometry as an intermediate representation to bridge the simulator and the generative model. This approach enhances controllability, minimizes adaptation cost through early alignment with real-world distributions, and supports off-the-shelf use of existing and future pretrained generative models. We further construct a D3Sim dataset to facilitate the training and evaluation of hybrid generation pipelines. Experiments demonstrate that Dreamland outperforms existing baselines with 50.8% improved image quality, 17.9% stronger controllability, and has great potential to enhance embodied agent training. Code and data will be made available.