Research map / 3D vision / 3D generation
World2Minecraft: Occupancy-Driven simulated scenes Construction
Lechao Zhang, Haoran Xu, Jingyu Gong, Xuhong Wang, Yuan Xie, Xin Tan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add SFT on MinecraftVLN | Qwen2.5-VL-3B · MinecraftVLN Combined Next-View | accuracy: 0.2288 -> 0.5609 | Improved |
| add SFT on MinecraftVLN | Qwen2.5-VL-7B · MinecraftVLN Combined Next-View | accuracy: 0.2878 -> 0.6642 | Improved |
| use RFT instead of SFT | Qwen2.5-VL-3B · MinecraftVLN Combined Next-View | accuracy: 0.5609 -> 0.3137 | Hurt |
| use RFT instead of SFT | Qwen2.5-VL-7B · MinecraftVLN Combined Next-View | accuracy: 0.6642 -> 0.6753 | No clear effect |
| use RFT instead of SFT | Qwen2.5-VL-3B · MinecraftVLN Combined Next-Action | accuracy: 0.4835 -> 0.6570 | Improved |
| use RFT instead of SFT | Qwen2.5-VL-7B · MinecraftVLN Base Next-Action | accuracy: 0.8000 -> 0.6343 | Hurt |
| add MinecraftOcc 8k to NYUv2 training (ablation) | Symphonies · NYUv2 occupancy | IoU: 49.91 -> 50.34 | Improved |
| add MinecraftOcc 8k to NYUv2 training (ablation) | Symphonies · NYUv2 occupancy | mIoU: 29.70 -> 29.91 | Improved |
| scale MinecraftOcc training from 8k to 100k images | Monoscene · MinecraftOcc | IoU: 40.66 -> 29.23 | Hurt |
| use occupancy-based reconstruction instead of LayoutGPT | World2Minecraft · Scene layout comparison | Semantic integrity: 0.689 -> 0.913 | Improved |
| add manual refinement to automated reconstruction instead of building from scratch | World2Minecraft · Scene building time | seconds per scene: 482.00 -> 70.38 | Improved |
Closest papers by meaning
- AdaOcc: Adaptive 3D Occupancy Prediction for Embodied Tasks (NeurIPS 2026 · similarity 0.87)
- Habitat Synthetic Scenes Dataset (HSSD-200): An Analysis of 3D Scene Scale and Realism Tradeoffs for ObjectGoal Navigation (CVPR 2024 · similarity 0.87)
- SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent (NeurIPS 2025 · similarity 0.86)
- Learning 3D Persistent Embodied World Models (NeurIPS 2025 · similarity 0.86)
- 3D Occupancy Prediction with Low-Resolution Queries via Prototype-aware View Transformation (CVPR 2025 · similarity 0.86)
- OctOcc: High-Resolution 3D Occupancy Prediction with Octree (AAAI 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
