cs.CVNov 15, 2026

PanDWM: Panoramic 4D Driving World Models

Authors: Jiajun Sun, Zeyu Zhang, Zeyu Ren, Guangfeng Jiang, Yang Zhao, Hao Tang

Organizations: Shenzhen University · Peking University · The University of Melbourne · La Trobe University

Abstract

Driving world models usually generate videos from six pinhole cameras with separate optical centres and small overlaps, and must learn to keep these views consistent. We present PanDWM, which stitches them into one 360∘360^\circ panoramic band. Camera boundaries then fall inside one image, so keeping the views consistent means keeping the image continuous, and each pixel still records its camera ray. Conditioned on layouts, a causal student generates the band window by window in four steps; it is distilled from a causal teacher and trained on its own rollouts. A feed-forward decoder lifts each frame along these rays into 3D Gaussians, and static Gaussians from neighbouring frames complete them into 4D scenes. On nuScenes, scored in one stitched 360∘360^\circ view, PanDWM has lower FID and FVD than the multi-camera baselines at 88-46×46\times their speed, and keeps its 50-step teacher's FID while lowering FVD by 22-25%.

Figures & tables

Explore similar work

CardsList
  1. PanoWorld: Real-World Panoramic Generation

    Jul 10, 2026Haoyuan Li, Dizhe Zhang, Yuemei Zhou +7World Model LearningWorld Models

  2. MUGEN: Interactive Panoramic World Exploration via Camera Control

    Sep 29, 2026Jiaming Tan, Zhen Li, Shuwei Shi +6Camera-Controlled Video Generation