| Depth-Anything-V2-Base -- unofficial transformers-format package (trust_remote_code) |
| Packaged by bigshanedogg, 2026. NOT an official release of the original authors. |
|
|
| Components |
| ---------- |
| 1. Depth Anything V2 Base weights (depth_anything_v2_vitb.pth) |
| Source: https://huggingface.co/depth-anything/Depth-Anything-V2-Base |
| Copyright: Lihe Yang, Bingyi Kang, Zilong Huang, Zhen Zhao, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao |
| License: CC-BY-NC-4.0 (non-commercial) -- LICENSE |
| 2. Depth Anything V2 code (depth_anything_v2/ and metric_depth/depth_anything_v2/, incl. its DINOv2 copy; vendored into dav2_*.py) |
| Source: https://github.com/DepthAnything/Depth-Anything-V2 @ a561b849ebae10a6f5ef49e26c83cbbcd36c71bf |
| Copyright: the Depth Anything V2 authors; DINOv2 layers Copyright (c) Meta Platforms, Inc. and affiliates. |
| License: Apache-2.0 -- LICENSE.code |
|
|
| Changes made in this package |
| ---------------------------- |
| - dav2_*.py: the official files concatenated into flat files; package-internal imports replaced by relative imports; NestedTensorBlock spelled out where dinov2.py imported it as Block. No computation changed. |
| - configuration_ / modeling_ / image_processing_ / processing_depth_anything_v2.py: new transformers wrappers; the model runs the official DepthAnythingV2 methods (relative, or metric when max_depth is set); the image processor is image2tensor's transform. |
| - model.safetensors: the .pth state dict converted by save_pretrained, names unchanged, every tensor checked equal. |
|
|
| License files |
| ------------- |
| - LICENSE: Creative Commons Attribution-NonCommercial 4.0 International (weights) |
| - LICENSE.code: Apache License 2.0, copy from DepthAnything/Depth-Anything-V2 (code) |
|
|
