English(EN)4Director: Controlling Video World Models with Rigid 3D Geometry
新AI模型提供对三维视频生成和动力学的增强控制
作者PulseAugur 编辑部·[9 个来源]·
研究人员正在探索控制和理解视频生成的新方法。一种方法,Generative Cinematographer (GenCine),允许艺术家通过指定路径和使用局部三维运动手柄来创作三维摄像机和物体运动。另一项开发,4Director,使用具有重建网格的显式四维场景表示来实现对摄像机和物体运动的精确控制,优于先前的方法。此外,DynaTokens 提供了一种轻量级的方法,通过使用可学习的场景特定令牌来教授现有摄像机控制视频模型动力学,从而改善了动力学和摄像机控制之间的权衡。Reimagine Video Dynamics (RVD) 专注于通过将可编辑的动力学令牌与视觉上下文分离来编辑视频动力学,从而实现场景演变的直接操作。
AI
arXiv:2609.38877v1 Announce Type: new Abstract: Physical parameter estimation from video aims to recover the parameters of a known family of governing dynamical equations from pixel observations. Existing identifiability theory for this setting has focused on linear time-invarian…
arXiv:2610.02180v1 Announce Type: cross Abstract: Current controllable video generation systems often rely on 2D motion trajectories or sparse drag signals for object motion. These controls are ambiguous because the same 2D trajectory can correspond to different 3D motions, espec…
arXiv:2609.33935v2 Announce Type: replace-cross Abstract: Adapting video models to new tasks typically requires dedicated data curation and fine-tuning. While visual analogy provides a training-free alternative by specifying tasks in-context, it remains restricted to the image do…
arXiv cs.AI
TIER_1English(EN)·Ziqi Zhou, Yujian Yuan, Laura Sevilla-Lara·
arXiv:2609.38683v1 Announce Type: cross Abstract: Cinematic camera motion is a fundamental storytelling tool, defined not only by where the camera is positioned in the scene, but also by how it moves in terms of direction and speed. Recent work on camera trajectory generation and…
Precise control over camera and object motion is essential for professional video production. Existing methods control objects only coarsely, through image-plane cues that are ambiguous in depth and rotation or through 3D tracks and blobs that lack complete geometry and lose cons…
Video generation must account for two sources of motion, one induced by the observer's camera path and the other caused by scene dynamics. An ideal camera-controlled video model should account for both motions: let users move the camera while evolving the scene dynamics. While cu…
arXiv cs.CV
TIER_1English(EN)·Wei Cao, Hao Zhang, Vikram Voleti, Yuqun Wu, Mallikarjun B R, Shimon Vainer, Mark Boss, Yaoyao Liu·
arXiv:2610.02160v1 Announce Type: new Abstract: Precise control over camera and object motion is essential for professional video production. Existing methods control objects only coarsely, through image-plane cues that are ambiguous in depth and rotation or through 3D tracks and…
arXiv cs.CV
TIER_1Italiano(IT)·Yu Yuan, Yawen Lu, Guoxian Song, Kevin Duarte, Ratheesh Kalarot, Di Chang, Xijun Wang, Stanley H. Chan·
arXiv:2609.36496v1 Announce Type: new Abstract: Most video editing methods focus on changing the appearance of the source video, while offering limited control over its dynamics. We introduce Reimagine Video Dynamics (RVD), a framework that disentangles a compact, editable dynami…
SOMNIA: Reconstructing Renderer State From AI Video An operational experiment in recovering temporal rendering guidance from synthetic footage that never had a renderer https://youtu.be/PGwVhuuKol0 On October 3, 2026, Black Sail Studio completed the first end-to-end run of SOMNIA…