Skip to content

论文HF Daily Papers · arXiv 2611.244084 天前

StrideWorld基准发布:量化视频世界模型场景刚性与动作一致性缺陷 ​

本文由 AI 根据公开内容整理,可能有误,请以原文为准:StrideWorld: Benchmarking Scene Rigidity and Action Consistency in Interactive Video World Model(HF Daily Papers · arXiv 2611.24408)

核心内容 ​

针对现有视频世界模型基准依赖自适应参考系导致误差被掩盖的问题,研究团队推出StrideWorld基准。该基准基于起始视图真值几何评估场景刚性,并利用单指令步长(stride)作为自参照指标检测动作一致性。在140个起始视图的开源模型测试中,除一个模型外均出现显著场景变形,偏移量最高达刚性渲染的7.3倍;组合路径下多数模型终点偏离自身步长预测位置,验证了该基准与人类判断的高度一致性。

技术方案上,StrideWorld通过测量模型初始步长确立后续动作的一致性判据,避免自由尺度对齐带来的评估失真。实验表明,仅有两类引入估计场景深度进行指令缩放的模型能使组合路径接近预测位置。此外,利用少量渲染视频片段对两个模型进行微调,有效消除了其在复杂路径中的中途停滞现象。相关代码、数据集及排行榜已在GitHub开源,为具身智能仿真训练提供了更严苛的视觉生成质量度量工具。

关键看点 ​

  • 评测机制创新:摒弃拟合测试视频的自适应参考系,改用起始视图真值几何与模型自身初始步长作为固定基准,精准检出渐进式场景变形与节奏漂移。
  • 量化变形缺陷:在140个起始视图测试中,几乎所有开源模型随相机移动发生场景变形,最大几何偏移量达到同路径刚性渲染结果的7.3倍。
  • 深度缩放关键性:仅有两个采用估计场景深度对指令进行缩放的模型,在组合路径任务中实现了终点位置与自身步长预测的高精度对齐。
  • 微调修复停滞:实验证实使用少量渲染视频片段对模型进行微调,可有效解决生成视频在复杂组合路径执行过程中频繁出现的中途停滞问题。

工程与产业观察 ​

该基准揭示了当前视频世界模型在3D结构保持与长时序动作执行上的共性短板,特别是缺乏深度感知导致的轨迹漂移。对于具身智能而言,高保真的物理一致性是Sim-to-Real迁移的前提,StrideWorld提供的刚性约束评测与步长自校验方法,为筛选合格的世界模型底座及优化仿真数据生成管线确立了量化标准。

英文摘要 (Abstract) ​

Interactive video world models turn user commands into the exploration of a generated scene, which requires the scene to keep its 3D structure as the camera moves and each action to keep a stable effect. Existing benchmarks mostly evaluate these properties against references fitted to the video under test, either a scene reconstructed from its own frames or a camera trajectory aligned to the command path with a free scale. Such references can absorb gradual scene deformation and changes in pace and route length, so these errors can pass undetected. We therefore introduce StrideWorld, a benchmark whose references are fixed before the motion they judge. It evaluates scene rigidity against the ground-truth geometry of the start view, and action consistency against the stride of each action, the camera motion of one command. The stride is measured on the model's first steps and then used to test the steps that follow, so each model is judged only against its own steps. Its measurements are validated on rendered videos with known camera motion, and its scores agree with human judgments. Across videos generated by open models from 140 start views, all models but one deform their scenes as the camera moves, with offsets up to 7.3 times those of a rigid rendering along the same camera path. Most models keep a steady pace when a single action is held. Along composed routes, several models often stall, and most completed routes end away from the positions predicted by each model's own strides. Only the two models that scale commands by the estimated scene depth bring most routes close to these positions. Fine-tuning two models on a few rendered clips removes most of their mid-route stalls. Code, data, and a leaderboard are publicly available at https://github.com/chuangwang123/strideworld.

Hugging Face Daily Papers 热度:11 upvotes 作者:Chuang Wang, Rui Nie, Jie Song, Junyuan Gao, Jiahe Song, Haitao Zhou, Boyu Zhu, Xing Li, Jiang Wu, Conghui He, Qian Yu, Weijia Li

基于本页技术方向关联的整机参数、核心零部件、实测报告与同赛道选型横评。

外部来源

本页事实对应的公开页面,以原文为准。(点击将离开本站)

  1. HF Daily Papers · arXiv 2611.24408:StrideWorld: Benchmarking Scene Rigidity and Action Consistency in Interactive Video World Model ↗访问 2026-11-20