OneSteer: Unlocking Steerable General Humanoid Motion Tracking via One-Step Diffusion Whole-Body Control
arXiv 收到一篇题为 OneSteer 的替换版论文(编号 arXiv:2610.41354v4),研究通过一步扩散全身控制实现可操控的通用人形运动跟踪。同一批投稿中还出现多篇相关人形运动跟踪与全身控制论文,包括 MoViC(从单目视频流实时全身控制)以及 XMimic 与 XSoma(不到一分钟训练即可学习人形运动跟踪与崎岖地形移动)。这些条目目前均为摘要级别的发布信息,具体方法结果与实验数据尚未在材料中披露。
How this story unfolded
Key points8 items Model summary · ordered by when the evidence was published
- Point 01 11/16, 13:00
OneSteer 以替换版形式发布,编号 arXiv:2610.41354v4。
Based on 1 - Point 02 11/16, 13:00
OneSteer 面向无需修订轨迹、仅指定部署目标的可操控人形运动跟踪。
Based on 1 - Point 04 11/18, 13:00
MoViC 从单目视频流进行实时人形全身模仿控制。
Based on 1 - Point 05 11/18, 13:00
MoViC 需要稳健的通用跟踪器与高质量实时重建。
Based on 1 - Point 06 11/18, 13:00
XMimic 与 XSoma 在不到一分钟训练内学习人形运动跟踪。
Based on 1 - Point 07 11/18, 13:00
XMimic 与 XSoma 同时学习崎岖地形上的移动。
Based on 1 - Point 08 11/18, 13:00
PromptPack 针对扩散编辑器提出免训练的意图对齐空间与语义控制增强。
Based on 1
ReportsBy publish time · newer further down
- November 16, 2026 · Mon
- Start
OneSteer: Unlocking Steerable General Humanoid Motion Tracking via One-Step Diffusion Whole-Body Control
Original summary arXiv:2610.41354v4 Announce Type: replace Abstract: Deployment objectives often specify desired outcomes without providing revised trajectories, while parts of the original referen
arXiv cs.RO Original ↗ - November 18, 2026 · Wed
- +2d 0h
MoViC: Real-Time Humanoid Whole-Body Control from Monocular Video Streams
Original summary arXiv:2611.19998v1 Announce Type: new Abstract: Real-time whole-body imitation from monocular video requires both a robust universal tracker and high-quality real-time reconstructi
arXiv cs.RO Original ↗ - +2d 0h
PromptPack: Training-Free Augmentation of Diffusion Editors for Intent-Aligned Spatial and Semantic Control
Original summary arXiv:2611.21865v1 Announce Type: new Abstract: Many image-editing tasks require confining changes to intended spatial regions while exactly preserving user-designated source conte
arXiv cs.LG Original ↗ - +2d 0h
XMimic and XSoma: Learning Humanoid Motion Tracking and Locomotion in Under a Minute
Original summary arXiv:2611.20452v1 Announce Type: new Abstract: We present XMimic and XSoma, which learn humanoid motion tracking and locomotion on rough terrain in under a minute of training. Bot
arXiv cs.RO Original ↗ - Still developing · later reports will keep extending this axis