ps-core-models-apache-2.0

ONNX model weights re-hosted for the DepthField Photo desktop app, which downloads them at runtime. All files here are Apache-2.0 and are redistributed unmodified from their original sources. This repo is a stable mirror so the app fetches from a source we control; integrity is pinned by sha256 in the app's manifest.

Contents & attribution

l1_encoder.onnx, l1_decoder.onnx — EfficientViT-SAM (L1)

• Original author: MIT Han Lab
• Source: https://github.com/mit-han-lab/efficientvit (ONNX export originally published at https://huggingface.co/mit-han-lab/efficientvit-sam)
• License: Apache License 2.0 — https://www.apache.org/licenses/LICENSE-2.0
• Modifications: none.

depth_anything_v2_small.onnx — Depth Anything V2 (Small)

• Original authors: Depth Anything V2 authors; ONNX conversion by onnx-community
• Source: https://github.com/DepthAnything/Depth-Anything-V2 (ONNX originally at https://huggingface.co/onnx-community/depth-anything-v2-small)
• License: Apache License 2.0 — https://www.apache.org/licenses/LICENSE-2.0 (the Small checkpoint is Apache-2.0; Base/Large are CC-BY-NC and are NOT included here)
• Modifications: none.

License

Licensed under the Apache License, Version 2.0. A copy of the License is available at https://www.apache.org/licenses/LICENSE-2.0. Distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND.