English

EvoWorld: Evolving Panoramic World Generation with Explicit 3D Memory

Computer Vision and Pattern Recognition 2025-10-02 v1

Abstract

Humans possess a remarkable ability to mentally explore and replay 3D environments they have previously experienced. Inspired by this mental process, we present EvoWorld: a world model that bridges panoramic video generation with evolving 3D memory to enable spatially consistent long-horizon exploration. Given a single panoramic image as input, EvoWorld first generates future video frames by leveraging a video generator with fine-grained view control, then evolves the scene's 3D reconstruction using a feedforward plug-and-play transformer, and finally synthesizes futures by conditioning on geometric reprojections from this evolving explicit 3D memory. Unlike prior state-of-the-arts that synthesize videos only, our key insight lies in exploiting this evolving 3D reconstruction as explicit spatial guidance for the video generation process, projecting the reconstructed geometry onto target viewpoints to provide rich spatial cues that significantly enhance both visual realism and geometric consistency. To evaluate long-range exploration capabilities, we introduce the first comprehensive benchmark spanning synthetic outdoor environments, Habitat indoor scenes, and challenging real-world scenarios, with particular emphasis on loop-closure detection and spatial coherence over extended trajectories. Extensive experiments demonstrate that our evolving 3D memory substantially improves visual fidelity and maintains spatial scene coherence compared to existing approaches, representing a significant advance toward long-horizon spatially consistent world modeling.

Keywords

Cite

@article{arxiv.2510.01183,
  title  = {EvoWorld: Evolving Panoramic World Generation with Explicit 3D Memory},
  author = {Jiahao Wang and Luoxin Ye and TaiMing Lu and Junfei Xiao and Jiahan Zhang and Yuxiang Guo and Xijun Liu and Rama Chellappa and Cheng Peng and Alan Yuille and Jieneng Chen},
  journal= {arXiv preprint arXiv:2510.01183},
  year   = {2025}
}

Comments

Code available at: https://github.com/JiahaoPlus/EvoWorld

R2 v1 2026-07-01T06:11:18.314Z