PulseAugur
中
实时 11:10:00
English(EN) Splat2Act: From Captured Scenes to Personalized VLA Manipulation

Splat2Act 管道使用 3D 场景重建来个性化机器人操作策略

研究人员开发了 Splat2Act,这是一个新颖的管道,可将通用的视觉-语言-动作 (VLA) 操作策略个性化,以适应特定环境和物体。该系统使用手持手机扫描,通过 3D 高斯飞溅和碰撞几何来创建场景的逼真数字孪生。然后,它会自动识别容器,生成拾取任务,并使用模拟神谕创建演示,仅需要人类进行场景和物体捕捉。当应用于 101 个校园环境时,该管道生成了超过 68,000 个拾取演示,并且微调后的策略在目标场景中达到了 62.5% 的拾取成功率,证明了 VLA 操作的自动化从现实到模拟再到现实训练的潜力。 AI

影响 以更低的数据收集成本,实现现实环境中个性化的机器人操作。

排序理由 该项目是一篇研究论文,详细介绍了一种新的机器人操作方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

Splat2Act 管道使用 3D 场景重建来个性化机器人操作策略

本文如何被排名

Signal score
10 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该项目是一篇研究论文,详细介绍了一种新的机器人操作方法。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
product, infra, paper
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Same-day
Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.CV TIER_1 English(EN) · Xiaomeng Ye, Krishna Jaganathan, Gyubeum Lim, Gunjan Chhablani, Muhammad Zubair Irshad, Zsolt Kira ·

    Splat2Act:从捕捉场景到个性化VLA操控

    arXiv:2610.13945v1 Announce Type: cross Abstract: A user who brings a generalist manipulation policy into their home or office wants it to pick up their objects from their shelves, counters, and tables. Yet even a strong generalist like $\pi_{0.5}$ achieves 0% pickup success in o…