NOTICE

Camera-Depth-Model-D405 -- unofficial transformers-format package (trust_remote_code)
Packaged by bigshanedogg, 2026. NOT an official release of the original authors.

Components
----------
1. Camera Depth Model weights (cdm_d405.ckpt)
   Source: https://huggingface.co/depth-anything/camera-depth-model-d405
   Copyright: Minghuan Liu, Zhengbang Zhu, Xiaoshen Han, Peng Hu, Haotong Lin, Xinyao Li, Jingxiao Chen, Jiafeng Xu, Yichu Yang, Yunfeng Lin, Xinghang Li, Yong Yu, Weinan Zhang, Tao Kong, Bingyi Kang (ByteDance Seed)
   License: CC-BY-NC-4.0 (non-commercial) -- LICENSE
2. Camera Depth Model code (cdm/rgbddepth/ incl. its DINOv2 copy, vendored into cdm_*.py; cdm/infer.py load_images / load_model steps in the image processor and build)
   Source: https://github.com/ByteDance-Seed/manip-as-in-sim-suite @ 757b2ea26bcf7992c92ac80485d82b905e961850
   Copyright: Copyright (c) 2025 Bytedance Ltd. and/or its affiliates; DINOv2 layers Copyright (c) Meta Platforms, Inc. and affiliates.
   License: Apache-2.0 -- LICENSE.code

Changes made in this package
----------------------------
- cdm_*.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_camera_depth_model.py: new transformers wrappers; the model runs RGBDDepth.forward / image2tensor; the image processor is infer.py load_images + image2tensor.
- model.safetensors: the checkpoint's model state dict (prefix removed as infer.py load_model), converted by save_pretrained, every tensor checked equal. The Lightning checkpoint's training state and its unused _mean / _std (ImageNet statistics; the original strict=False load ignores them) dropped.

License files
-------------
- LICENSE: Creative Commons Attribution-NonCommercial 4.0 International (weights)
- LICENSE.code: Apache License 2.0, copy from ByteDance-Seed/manip-as-in-sim-suite (code)