Datasets:
Cross-link released models (pretrained + NYU-FT)
Browse files
README.md
CHANGED
|
@@ -265,6 +265,13 @@ The training split (17,799 frames) spans 10 games; no single game dominates. The
|
|
| 265 |
Full corpus before session-capping and the train split is 48,615 frames. Machine-readable counts in
|
| 266 |
[`results/game_distribution.json`](results/game_distribution.json).
|
| 267 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 268 |
## Methodology and scope
|
| 269 |
|
| 270 |
Full detail in [`METHODOLOGY.md`](METHODOLOGY.md); machine-readable metrics (including per-frame) in
|
|
|
|
| 265 |
Full corpus before session-capping and the train split is 48,615 frames. Machine-readable counts in
|
| 266 |
[`results/game_distribution.json`](results/game_distribution.json).
|
| 267 |
|
| 268 |
+
## Released models
|
| 269 |
+
|
| 270 |
+
Trained with this dataset and released under the same license (satisfying the model-release clause):
|
| 271 |
+
|
| 272 |
+
- [`originlab/lotus-game-depth-pretrained`](https://huggingface.co/originlab/lotus-game-depth-pretrained) - pre-trained from scratch on the game depth; zero-shot KITTI AbsRel 0.191.
|
| 273 |
+
- [`originlab/lotus-game-depth-nyu-ft`](https://huggingface.co/originlab/lotus-game-depth-nyu-ft) - the above fine-tuned on NYU; NYU AbsRel 0.116.
|
| 274 |
+
|
| 275 |
## Methodology and scope
|
| 276 |
|
| 277 |
Full detail in [`METHODOLOGY.md`](METHODOLOGY.md); machine-readable metrics (including per-frame) in
|