Genie Sim PanoRecon: Fast Immersive Scene Generation from Single-View Panorama

TL;DR

Genie Sim PanoRecon is a fast, feed-forward pipeline that creates high-fidelity 3D scenes from single panoramas for robotic simulation, using a novel depth-aware fusion and depth-injection for geometric consistency.

Contribution

It introduces a novel, training-free depth-injection module and a parallel processing pipeline for coherent 3D scene reconstruction from panoramas.

Findings

Reconstructs photo-realistic scenes in seconds.

Integrates seamlessly into Genie Sim for synthetic data generation.

Achieves geometric consistency across views.

Abstract

We present Genie Sim PanoRecon, a feed-forward Gaussian-splatting pipeline that delivers high-fidelity, low-cost 3D scenes for robotic manipulation simulation. The panorama input is decomposed into six non-overlapping cube-map faces, processed in parallel, and seamlessly reassembled. To guarantee geometric consistency across views, we devise a depth-aware fusion strategy coupled with a training-free depth-injection module that steers the monocular feed-forward network to generate coherent 3D Gaussians. The whole system reconstructs photo-realistic scenes in seconds and has been integrated into Genie Sim - a LLM-driven simulation platform for embodied synthetic data generation and evaluation - to provide scalable backgrounds for manipulation tasks. For code details, please refer to: https://github.com/AgibotTech/genie_sim/tree/main/source/geniesim_world.