Downloads · 30 days
10
24% of all-time downloads
roomscan-dav2-metric-large-ft-lidar
NackPanupong/roomscan-dav2-metric-large-ft-lidar
roomscan-dav2-metric-large-ft-lidar is a depth estimation model from NackPanupong. Use it for the depth estimation task on the model card, and read the license before you ship it in a product. It is set up for transformers.
Depth Anything V2 Metric-Indoor fine-tuned on ARKitScenes (roomscan, ADR-013): frozen DINOv2 backbone, DPT neck + head trained with SiLog on the lidar teacher.
All-time downloads
41
Public
Parameters
335M
1.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.3 GB · 100%
At a glance
- Task
- Depth Estimation
- Library
- transformers
- Model type
- depth_anything
- Access
- Public
- Created
- Sep 21, 2026
- Updated
- Sep 21, 2026
- SHA
- d4e8a48e
Base models
Model card
From the Hugging Face model README
roomscan-dav2-metric-large-ft-lidar
Depth Anything V2 Metric-Indoor fine-tuned on ARKitScenes (roomscan, ADR-013): frozen DINOv2 backbone, DPT neck + head trained with SiLog on the lidar teacher.
| run | r2_ft_lidar (best/, step 2000) |
| train sets | ['train_faro'] — 10 scenes, 8656 frames |
| val (ARKitScenes Validation fold, Faro) | abs_rel 0.14224483345558936 (pretrained step 0: 0.46101968864375237) |
| config sha1 | 03437036f19d |
Test scenes (the six Exp1 scenes) were never seen during training, validation or checkpoint choice.
- Task
- Depth Estimation
- Library
- transformers
- Type
- depth_anything
- Created
- Sep 21, 2026
- Updated
- Sep 21, 2026