naver/blast3r-depth

BLASt3R — depth model

Multi-channel depth model for BLASt3R, a method for bundle-adjusted 3D reconstruction and visual SLAM from image collections and video. Code: github.com/naver/blast3r.

This is one of the two checkpoints BLASt3R needs. The other is the matching model, naver/blast3r-matcher.

Training of this checkpoint was initialized from DINOv3 pretrained weights, so these weights are a derivative work of DINOv3 — see License below.

Contents

config.json          model class and its arguments
model.safetensors    the weights

config.json names the class and hyperparameters, so loading a checkpoint never unpickles anything.

Usage

from huggingface_hub import snapshot_download

ckpt_match = snapshot_download('naver/blast3r-matcher')
ckpt_depth = snapshot_download('naver/blast3r-depth')

Then run reconstruction from the BLASt3R code repository:

python scripts/run_offline.py \
  --images /path/to/images \
  --output /path/to/output \
  --ckpt_match "$CKPT_MATCH" \
  --ckpt_depth "$CKPT_DEPTH" \
  --mono_cam

License

Released under the NAVER non-commercial license (LICENSE), which covers the code and these checkpoints. Third-party subcomponents, the datasets these weights were trained on, and the pretrained weights they were initialized from are acknowledged in NOTICE; dataset terms pass through to the weights.

This checkpoint was initialized from DINOv3 weights, made available by Meta Platforms, Inc. under the DINOv3 License. Its use is therefore additionally subject to that license, including its restrictions on military, warfare, nuclear, espionage and weapons-related use and on activities subject to Trade Controls. As Section 1.b of that license requires, a copy is distributed here as DINOv3_LICENSE.md.