Search papers, labs, and topics across Lattice.
This paper presents Genie Sim PanoWorld, a novel two-stage feed-forward pipeline that generates high-fidelity 3D scenes from a single 360-degree panorama without the need for per-scene optimization or multi-view capture. By integrating a trajectory-controllable panoramic video with a latent video diffusion model and employing a self-consistency objective, the method achieves superior performance in generating panoramic videos and reconstructing 3D scenes. The results demonstrate that Genie Sim PanoWorld not only outperforms existing geometry-conditioned baselines but also generalizes effectively to unseen indoor environments, making it a valuable asset for embodied AI applications.
Achieving high-fidelity 3D scene reconstruction from a single panorama could revolutionize the way we create and interact with virtual environments.
We address the problem of reconstructing a high-fidelity, freely navigable 3D scene from a single $360^\circ$ panorama, without per-scene optimization or multi-view capture. Existing methods either lack metric trajectory control, which hinders reliable downstream 3D reconstruction, or struggle with large disocclusions under long-range camera motion while requiring high-end multi-GPU servers.We present Genie Sim PanoWorld, a two-stage feed-forward pipeline that bridges generation and reconstruction via an explicit, trajectory-controllable panoramic video. A NavMesh-planned $\mathrm{SE}(3)$ roaming trajectory is injected into a latent video diffusion model through dense geometry-warped conditioning; long--short trajectory mixed training and a self-consistency objective based on shortcut models together yield high-fidelity video in four CFG-free denoising steps. A feed-forward panoramic reconstructor then lifts the generated video into a high-fidelity 3D Gaussian scene that supports real-time, free-viewpoint roaming and can be directly used as a simulation-ready asset for embodied AI applications. Experiments show that Genie Sim PanoWorld outperforms geometry-conditioned baselines in both panoramic video generation and downstream 3D reconstruction, while generalizing zero-shot to unseen indoor scenes.
