PulseAugur
实时 04:35:49

SceneCode generates editable 3D scenes from text prompts

Researchers have developed SceneCode, a framework that generates editable indoor scenes by compiling natural language prompts into executable Python programs for Blender. This approach moves beyond static meshes, creating physically interactable assets with simulation-ready articulation metadata. The system uses an agentic backbone and a planner-designer-critic loop to generate and refine object-specific code, enabling traceable and locally editable scene construction. AI

影响 Enables more controllable and interactive 3D scene generation for embodied AI and robotics.

排序理由 The cluster describes a new research paper detailing a novel framework for scene generation. [lever_c_demoted from research: ic=1 ai=1.0]

在 Hugging Face Daily Papers 阅读 →

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

报道来源 [1]

  1. Hugging Face Daily Papers TIER_1 English(EN) ·

    SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objects

    Indoor scene synthesis underpins embodied AI, robotic manipulation, and simulation-based policy evaluation, where a useful scene must specify not only what the environment looks like, but also how its objects are structured. Existing pipelines, however, typically represent genera…