| # MetricScenes |
|
|
| A metrically-grounded, in-the-wild dataset. For more details, please visit the [project page](https://metricscenes.github.io/). |
|
|
| ## Paper |
|
|
| **Title:** Honey, I Shrunk the Arc de Triomphe! |
| **Authors:** Yuanbo Xiangli, Hanyu Chen, Xueqing Tsang, Noah Snavely |
| **Project page:** https://metricscenes.github.io/ |
|
|
| ### Abstract |
|
|
| Metric scale monocular geometry estimation has seen significant progress through large-scale data aggregation, yet current foundation models suffer from a persistent ''scale-collapse'' phenomenon: distant landmarks and vast landscapes are metrically underestimated. This performance gap stems from a training data bottleneck, where existing metric-scale datasets are hardware-constrained to unvaried street-level LiDAR or short-range indoor scans, or consist of synthetic data that lacks the semantic complexity of the physical world. To bridge this gap, we curate a new metrically-grounded, in-the-wild dataset that we call Metricscenes, gathered from a variety of sources including Internet photo collections and stereo imagery. We estimate camera poses and initial depth maps for each scene using off-the-shelf methods, and recover absolute scale from geo-tagged metadata as well as known stereo camera baselines. We also improve the quality of depth maps derived from MetricScenes via a new two-stage Poisson completion method. Fine-tuning MoGe-2 on our dataset significantly mitigates scale-collapse and achieves superior metric accuracy in unconstrained, open-domain scenes while maintaining state-of-the-art performance on standard benchmarks. |
| ## Release Structure |
|
|
| MetricScenes is aggregated from AerialMegaDepth, MegaScenes, and Stereo4D. We develop pipelines to extract metric-scale depth maps in each case. |
| The public release is organized as dataset_name/scene_id/frame_id/...: |
| |
| ```text |
| MetricScenes/ |
| ├── AerialMegaDepth/ |
| │ ├── 0000 |
| │ │ ├── 1000570923_c2a177031b_o |
| │ │ │ ├── depth_complete.png |
| │ │ │ ├── depth_partial.png |
| │ │ │ ├── image.jpg |
| │ │ │ └── meta.json |
| │ │ ├── 1001414672_f286cdb145_o |
| │ │ │ └── ... |
| │ │ └── ... |
| │ ├── 0001 |
| │ │ └── ... |
| │ └── ... |
| ├── MegaScenes/ |
| │ ├── 000 |
| │ │ ├── 000352 |
| │ │ │ ├── depth_complete.png |
| │ │ │ ├── depth_partial.png |
| │ │ │ ├── image.jpg |
| │ │ │ └── meta.json |
| │ │ ├── 000373 |
| │ │ │ └── ... |
| │ │ └── ... |
| │ ├── 001 |
| │ │ └── ... |
| │ └── ... |
| ├── Stereo4D/ |
| │ ├── -3Sx43OYGJ8 |
| │ │ ├── 15081748_f99 |
| │ │ │ ├── depth_complete.png |
| │ │ │ ├── depth_partial.png |
| │ │ │ ├── image.jpg |
| │ │ │ └── meta.json |
| │ │ ├── 21755088_f99 |
| │ │ │ └── ... |
| │ │ └── ... |
| │ ├── -5JaYFNtYlM |
| │ │ └── ... |
| │ └── ... |
| │ |
| └── README.md |
| ``` |
| |
| depth_partial.png is the incomplete depth from SfM/MVS or off-the-shelf geometric models; |
| depth_complete.png is the completed depth using our proposed two-stage edge-aware Poisson completion method; |
| image.jpg is the RGB image; |
| meta.json contains camera parameters like intrinsics, extrinsics etc. |
| |
| ## Licensing Metadata |
| |
| The MetricScenes dataset is licensed under the Creative Commons Attribution 4.0 International License. **The original images come with their own licenses.** |
| |
| |
