cs.CVNov 16, 2026

Every Camera Grows Up: Camera-Height Regression in Interactive Video World Models

Authors: Sheng Lu, Guangpin Tao, Tong Lu, Ying Tai

Organizations: State Key Laboratory for Novel Software Technology, Nanjing University

Abstract

Interactive video world models serve as simulators for embodied agents: from a starting image and an action trajectory, they generate the views the agent would see. The starting image fixes how high the camera is, for example 0.15 m for a floor robot or 2.2 m for an overhead camera, and a faithful simulator must keep this height as the camera moves. We show that six recent open-weight models fail to do so: as motion replaces the starting view with generated content, they move the camera toward human eye level, a failure we call camera-height regression. Our benchmark, EyeHeight, pairs six rendered elevations (0.15 m to 2.2 m) and four calibrated real-world rig heights with scale-free multi-view height estimation. All six models lose their initial height during a 180∘180^\circ turn-and-walk rollout: initialized at 0.15 m, only 0-20% of clips end within 1.3×1.3\times of the starting height as inferred viewpoints converge to human heights of 1.1 m to 1.5 m, whereas ground-truth renders remain within a few percent. In Matrix-Game models, homography-tracked floor motion stays consistent with constant elevation during rotation even as newly outpainted room geometry yields a higher reconstructed viewpoint. Regression grows with rotation angle, begins while the start frame remains in context, and largely reverses in SANA-WM when a 360∘360^\circ turn restores the initial view. We trace this pull to human-centric video priors in which floor-level views are nearly absent: in SANA-WM, textual height cues and dense floor-disparity maps do not prevent it, whereas classifier-free text guidance during the first half of denoising strengthens it. Combining unguided sampling with height-balanced fine-tuning recovers more retention but costs 5 MUSIQ points, a loss fixed during those same early steps. Viewpoint drift alone shifts the paths predicted by visual navigation policies.

Figures & tables

Explore similar work

CardsList
  1. World in World: Explore the World with World Models

    Sep 10, 2026Chenxi Song, Yanming Yang, Chi ZhangCamera-Controlled Video GenerationLong-Horizon Video Generation

  2. WBench: A Comprehensive Multi-turn Benchmark for Interactive Video World Model Evaluation

    May 25, 2026Kaining Ying, Hengrui Hu, Siyu Ren +6World Model EvaluationInteractive World Models

  3. WorldMark: A Unified Benchmark Suite for Interactive Video World Models

    Apr 23, 2026Xiaojie Xu, Zhengyuan Lin, Kang He +5Action-Conditioned World ModelsWorld Model Evaluation