Research map / 3D vision / Gaussian splatting
Wanderland: Geometrically Grounded Simulation for Open-World Embodied AI
Xinhao Liu, Jiaqi Li, Youming Deng, Ruxin Chen, Yingjia Zhang, Yifei Ma, Li Guo, Yiming Li, Jing Zhang, Chen Feng
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 |
|---|---|---|---|
| build simulation from LIV-SLAM geometry (poses + LiDAR init) instead of vision-only reconstruction | 3DGS · Wanderland NVS (interpolated views) | PSNR: 18.27 -> 20.37 | Improved |
| build simulation from LIV-SLAM geometry instead of COLMAP | 3DGS · Wanderland NVS (extrapolated views) | PSNR: 17.00 -> 17.92 | Improved |
| LIV-SLAM geometry vs Vid2Sim on extrapolated LPIPS | Wanderland · Wanderland NVS (extrapolated views) | LPIPS: 0.371 vs 0.445 | Hurt |
| RL post-training in Vid2Sim-built environment | CityWalker · Wanderland test split | SR: SR -21% | Hurt |
| RL post-training in Wanderland geometrically grounded environment | CityWalker · Wanderland test split | SR: SR +14% | Improved |
| RL post-training in Wanderland environment | NoMaD · Wanderland test split | SR: SR +8% | Improved |
| RL post-training in Vid2Sim environment | NoMaD · Wanderland test split | SR: SR 0.24 (0%) | No clear effect |
| evaluate in Vid2Sim environments instead of Wanderland | navigation policies · Wanderland test split | SR | Hurt |
| use best vision-only method vs LIV-SLAM poses | COLMAP / feed-forward 3D models · Wanderland pose estimation | T-ATE scaled (m): 30 cm / 5 deg best-of-all error | Hurt |
| outdoor vs indoor scenes | navigation models · Wanderland benchmark | SR: NaVILA 0.47 -> 0.31 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Vid2Sim — Vid2Sim: Realistic and Interactive Simulation from Video for Urban Navigation (CVPR 2025) · better on 3 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator (NeurIPS 2026 · similarity 0.88)
- InternScenes: A Large-scale Simulatable Indoor Scene Dataset with Realistic Layouts (NeurIPS 2025 · similarity 0.88)
- Habitat Synthetic Scenes Dataset (HSSD-200): An Analysis of 3D Scene Scale and Realism Tradeoffs for ObjectGoal Navigation (CVPR 2024 · similarity 0.88)
- Synthesizing Photorealistic and Dynamic Urban Environments for Multimodal Robot Navigation and Collaboration (NeurIPS 2025 · similarity 0.87)
- SimVS: Simulating World Inconsistencies for Robust View Synthesis (CVPR 2025 · similarity 0.86)
- Vid2Sim: Realistic and Interactive Simulation from Video for Urban Navigation (CVPR 2025 · 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.
