PulseAugur
实时 08:58:40

FunRec reconstructs interactive 3D scenes from egocentric videos

Researchers have developed FunRec, a novel method for creating interactive 3D digital twins of indoor environments from egocentric RGB-D videos. This approach operates on unconstrained human interaction sequences, automatically identifying articulated parts, estimating their motion, and reconstructing both static and dynamic geometry. FunRec achieves significant improvements in part segmentation and pose accuracy compared to existing methods, enabling applications such as simulation-ready mesh export and robot-scene interaction. AI

影响 Enables creation of simulation-ready 3D environments from real-world interaction data.

排序理由 This is a research paper detailing a new method for 3D scene reconstruction.

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 →

FunRec reconstructs interactive 3D scenes from egocentric videos

报道来源 [1]

  1. arXiv cs.CV TIER_1 English(EN) · Alexandros Delitzas, Chenyangguang Zhang, Alexey Gavryushin, Tommaso Di Mario, Boyang Sun, Rishabh Dabral, Leonidas Guibas, Christian Theobalt, Marc Pollefeys, Francis Engelmann, Daniel Barath ·

    FunRec: Reconstructing Functional 3D Scenes from Egocentric Interaction Videos

    arXiv:2604.05621v2 Announce Type: replace Abstract: We present FunRec, a method for reconstructing functional 3D digital twins of indoor scenes directly from egocentric RGB-D interaction videos. Unlike existing methods on articulated reconstruction, which rely on controlled setup…