研究人员开发了用于机器人操控的新型3D世界模型创建方法。DepthWorld利用基于Stable Video Diffusion的模型,在RGB和深度数据上进行训练,以生成一致的3D环境,优于以前仅使用RGB的模型。另外,PointWAM通过预测场景和手部动力学作为点轨迹来建模3D世界动作,能够有效地在人类演示视频上进行预训练,并在灵巧性任务上表现优于现有方法。
AI
arXiv:2610.08780v1 Announce Type: cross Abstract: World models offer a data-driven alternative to traditional simulators for robotics, with applications spanning policy evaluation, improvement, and planning. All of these uses depend on faithful 3D geometry, yet current video-base…
World models offer a data-driven alternative to traditional simulators for robotics, with applications spanning policy evaluation, improvement, and planning. All of these uses depend on faithful 3D geometry, yet current video-based world models are trained on RGB alone and produc…
World action models jointly learn to forecast world dynamics and predict robot actions, such that the learned internal world dynamics guide accurate actions. Existing approaches typically represent the world as RGB frames or latent counterparts while predicting actions as end-eff…
Long-horizon compositional manipulation has become increasingly important for real-world robot deployment, where a single task involves multiple coordinated subtasks. Existing world-action models (WAMs) jointly predict short-horizon visual futures and actions, but typically lack …
We study robotic in-context learning (ICL), an emerging paradigm that enables robots to infer and execute tasks from visual demonstrations. Despite its growing promise, the problem itself remains under-defined: a visual demonstration simultaneously conveys action trajectories, ob…
arXiv cs.CV
TIER_1English(EN)·Jongbin Lim, Taeyun Ha, Seongho Cha, Kanghyeon Cho, Mingi Choi, Subin Jeon, Jisoo Kim, Byungjun Kim, Hanbyul Joo·
arXiv:2604.14944v3 Announce Type: replace-cross Abstract: We present HRDexDB, a real-world 4D dexterous grasping dataset capturing 3D hand-object interaction trajectories over time across five embodiments. The dataset comprises 3.2K trials over 100 diverse objects. Using a synchr…
arXiv cs.CV
TIER_1English(EN)·Chunghyun Park, Beomjun Kim, Seungcheol Park, Heeseung Kwon, Yashu Shukla, Seunghoon Sim, Jinwoo Shin, Minsu Cho·
arXiv:2610.02840v1 Announce Type: cross Abstract: World action models jointly learn to forecast world dynamics and predict robot actions, such that the learned internal world dynamics guide accurate actions. Existing approaches typically represent the world as RGB frames or laten…