研究人员开发了 World2Minecraft 系统，该系统通过预测 3D 语义占用率将真实世界场景转换为结构化的 Minecraft 环境。这使得在这些重建场景中进行视觉语言导航等下游任务成为可能。为了提高重建质量，创建了一个新的数据采集管道和大规模 MinecraftOcc 数据集，其中包含来自 156 个室内场景的超过 100,000 张图像。此外，一个名为 Dream-Cubed 的独立项目引入了一个大型 Minecraft 世界数据集和经过数十亿个方块训练的模型，用于高效、可控地生成交互式 3D 环境。 AI
影响 为具身 AI 研究和 3D 生成模型提供了更现实和可定制的模拟环境。
排序理由 该集群包含两篇 arXiv 论文，详细介绍了在 Minecraft 中生成和重建 3D 环境的新方法和数据集。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
Embodied intelligence requires high-fidelity simulation environments to support perception and decision-making, yet existing platforms often suffer from data contamination and limited flexibility. To mitigate this, we propose World2Minecraft to convert real-world scenes into stru…
arXiv:2604.27578v1 Announce Type: new Abstract: Embodied intelligence requires high-fidelity simulation environments to support perception and decision-making, yet existing platforms often suffer from data contamination and limited flexibility. To mitigate this, we propose World2…
Embodied intelligence requires high-fidelity simulation environments to support perception and decision-making, yet existing platforms often suffer from data contamination and limited flexibility. To mitigate this, we propose World2Minecraft to convert real-world scenes into stru…
arXiv:2604.22847v1 Announce Type: new Abstract: We introduce Dream-Cubed, a large-scale dataset of Minecraft worlds at voxel resolution, and a family of models using cubes as powerful compositional units for efficient generation of interactive 3D environments. Dream-Cubed compris…