cs.CVNov 3, 2026

ReconDreamer#: Self-Evolving Closed-Loop Synergy of Reconstruction and Generation for Dynamic Driving Scenes

Authors: Songling Liu, Guosheng Zhao, Xiaofeng Wang, Chaojun Ni, Guan Huang, Jiwen Lu, Xingang Wang, Zheng Zhu

Organizations: CAS Engineering Laboratory for Intelligent Industrial Vision, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China · School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China · Department of Automation, Tsinghua University, Beijing 100084, China · Giga AI, Beijing, China

Abstract

Dynamic driving scene reconstruction has made remarkable strides in recent years by leveraging generative priors, particularly in the quality of novel trajectory rendering. Nevertheless, a significant gap remains between the generated data and real-world sensor observations, severely hindering further improvements in reconstruction fidelity. In this paper, we present ReconDreamer#, a self-evolving framework that synergizes reconstruction and generation for dynamic driving scenes, which substantially bridges the gap between synthesized novel trajectory data and recorded real-world observations. Within this self-evolving closed-loop framework, the generative and reconstruction models mutually reinforce each other through iterative refinement, ultimately yielding a significant enhancement in the 3D representation capability of the reconstruction model. Additionally, we introduce a geometry-prior-based adaptive ground model to further enhance the rendering quality of the reconstruction model. To validate the effectiveness of the proposed method, we conduct comprehensive evaluations on Waymo, nuScenes, PandaSet, and EUVS benchmarks. Specifically on Waymo, at a 6-m lateral shift, ReconDreamer# improves NTA-IoU and NTL-IoU by 8.8% and 3.0%, respectively, and reduces FID by 15.1% relative to ReconDreamer++.

Figures & tables

Explore similar work

CardsList
  1. Diffusion-guided Generalizable Enhancer for Urban Scene Reconstruction

    May 21, 2026Henry Che, Jingkang Wang, Yun Chen +33D Gaussian SplattingDiffusion Model Guidance

  2. PointForward: Feedforward Driving Reconstruction through Point-Aligned Representations

    May 12, 2026Cheng Chi, Xianqi Wang, Hongcheng Luo +93D ReconstructionMulti-View 3D Reconstruction

  3. IGRecon4D: Dynamic-Object-Aware Feed-Forward Reconstruction of Driving Scenes

    Nov 16, 2026Yifei Tong, Ziyang Huang, Yizhao Han +2Dynamic Scene Reconstruction4D Scene Understanding