Downloads · 30 days
0
kshitijrajsharma/dinov3-hot-buildings
dinov3-hot-buildings is a image segmentation model from kshitijrajsharma. Use it for the image segmentation task on the model card, and read the license before you ship it in a product. It is set up for pytorch. The card lists the license as other.
Binary building-footprint segmentation from very-high-resolution RGB aerial imagery. Frozen DINOv3-ViT-L/16 (LVD1689M) encoder, terratorch UperNet decoder, multi-task auxiliary heads (boundary + signed distance), wate…
Downloads · 30 days
0
Access
Public
Updated Nov 13, 2026
Repo size
8.7 GB
Likes
0
Public
Click a slice to open those files.
.ckpt2.4 GB · 58%
From the Hugging Face model README
Binary building-footprint segmentation from very-high-resolution RGB aerial imagery.
Frozen DINOv3-ViT-L/16 (LVD1689M) encoder, terratorch UperNet decoder, multi-task
auxiliary heads (boundary + signed distance), watershed instance separation.
Decoder trained on hotosm/vhr-building-segmentation (~58k chips, OAM + OSM, global).
The checkpoint here is the decoder-only state_dict (encoder weights are loaded
fresh from kshitijrajsharma/dinov3 at inference / finetune time, so they are
stripped from this artifact to save bandwidth).
Decoder fine-tuning on small downstream OAM + OSM labelled sets (~100-200 chips)
inside the HOTOSM fAIr platform, or
direct inference via the upstream dinov3-hot CLI.
| File | Size | What it is |
|---|---|---|
dinov3l_upernet_hot.ckpt | ~493 MB | PyTorch Lightning checkpoint, decoder + pyramid + aux + main heads. Encoder excluded. |
The fAIr dinov3_buildings pipeline (in
hotosm/fAIr-models)
re-downloads the DINOv3-ViT-L/16 encoder from
kshitijrajsharma/dinov3 and
loads the decoder state from this checkpoint with strict=False.
| Benchmark | Pixel IoU | Instance [email protected] | avg vertices | orthogonality |
|---|---|---|---|---|
| HF global test split (7236 tiles) | 0.441 | 0.262 | 4.91 | 0.71 |
| Banepa, Nepal (1536x1536 OAM scene) | 0.656 | 0.407 | 5.99 | 0.63 |
Shape stats are computed in each region's local UTM zone after the production DP + MBR-safe regularize step.
get_intermediate_layers(norm=True, reshape=True, return_class_token=False).LearnedInterpolateToPyramidal — ConvTranspose-based learned
upsampling producing a 4-level pyramid.UperNetDecoder, decoder_channels=512,
pool_scales=(1, 2, 3, 6).Native input size 256 x 256. Dataset normalisation
mean=[0.4297, 0.4002, 0.3433], std=[0.2056, 0.1674, 0.1599].
The model files in this repository are distributed under the
DINOv3 License (Meta, last updated 2025-08-19), as declared in the
fAIr-models STAC item.
A copy of the licence is in LICENSE.md. The fAIr-models code that trains and serves
these models is AGPL-3.0.
If you use this checkpoint, please cite DINOv3:
@article{oquab2025dinov3,
title={DINOv3},
author={Oquab, Maxime and others},
journal={arXiv preprint arXiv:2508.10104},
year={2025}
}