Lensmera Depth — Depth Anything V2 Small (ONNX)
Verbatim mirror of onnx-community/depth-anything-v2-small (onnx/model_fp16.onnx,
opset 14, dynamic input shape) for the Lensmera app's on-device depth-based
background blur (bokeh). No modifications (renamed to dav2s_fp16.onnx).
License note: Only the Small variant is Apache-2.0. Base / Large / Giant are CC-BY-NC-4.0 (non-commercial) and must NOT be used.
- Upstream ONNX: https://huggingface.co/onnx-community/depth-anything-v2-small (Apache-2.0)
- Original model: https://github.com/DepthAnything/Depth-Anything-V2 (Small = Apache-2.0)
- sha256: 2df6223f206b5164e21f664ace61dabeb9bb6a49b8b5a3e00510b4807d0f5b04
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support