https://github.com/PzySeere/MetaSpatial

MetaSpatial leverages reinforcement learning to enhance 3D spatial reasoning in vision-language models (VLMs), enabling more structured, realistic, and adaptive scene generation for applications in the metaverse, AR/VR, and game development.
https://github.com/PzySeere/MetaSpatial

Last synced: about 1 year ago
JSON representation

MetaSpatial leverages reinforcement learning to enhance 3D spatial reasoning in vision-language models (VLMs), enabling more structured, realistic, and adaptive scene generation for applications in the metaverse, AR/VR, and game development.

• Host: GitHub
• URL: https://github.com/PzySeere/MetaSpatial
• Owner: PzySeere
• License: apache-2.0
• Created: 2025-03-08T03:37:57.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2025-03-25T03:36:46.000Z (about 1 year ago)
• Last Synced: 2025-03-25T04:27:02.376Z (about 1 year ago)
• Language: Python
• Homepage:
• Size: 8.1 MB
• Stars: 62
• Watchers: 5
• Forks: 3
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• Awesome-RL-for-LRMs - PzySeere/MetaSpatial
• StarryDivineSky - PzySeere/MetaSpatial - 语言模型(VLMs)的3D空间推理能力。它旨在实现更结构化、更逼真、更具适应性的场景生成，适用于元宇宙、增强现实/虚拟现实(AR/VR)和游戏开发等领域。该项目通过强化学习，让VLMs更好地理解和生成3D空间关系，从而创造更沉浸式的体验。MetaSpatial的核心在于提升模型对空间信息的理解和利用，使其生成的场景更符合物理规律和用户预期。项目目标是为构建更真实的虚拟世界提供技术支持，并推动相关领域的发展。 (3D视觉生成重建 / 资源传输下载)
• Awesome-Multimodal-Reasoning - [🖥️Code - | GRPO | 3D spatial reasoning | (Model / Image MLLM)