replace depth anything v1 with v2 safetensors

README.md

@@ -1,65 +1,108 @@
 ---
+library_name: transformers
+library: transformers
 license: cc-by-nc-4.0
-
-language:
-- en
-pipeline_tag: depth-estimation
-library_name: depth-anything-v2
 tags:
 - depth
 - relative depth
+pipeline_tag: depth-estimation
+widget:
+- inference: false
 ---
 
-# Depth-Anything-V2-Large
+# Depth Anything V2 Base – Transformers Version
 
-## Introduction
 Depth Anything V2 is trained from 595K synthetic labeled images and 62M+ real unlabeled images, providing the most capable monocular depth estimation (MDE) model with the following features:
 - more fine-grained details than Depth Anything V1
 - more robust than Depth Anything V1 and SD-based models (e.g., Marigold, Geowizard)
 - more efficient (10x faster) and more lightweight than SD-based models
 - impressive fine-tuned performance with our pre-trained models
 
-## Installation
+This model checkpoint is compatible with the transformers library.
 
-```bash
-git clone https://huggingface.co/spaces/depth-anything/Depth-Anything-V2
-cd Depth-Anything-V2
-pip install -r requirements.txt
-```
+Depth Anything V2 was introduced in [the paper of the same name](https://arxiv.org/abs/2406.09414) by Lihe Yang et al. It uses the same architecture as the original Depth Anything release, but uses synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. The original Depth Anything model was introduced in the paper [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) by Lihe Yang et al., and was first released in [this repository](https://github.com/LiheYoung/Depth-Anything).
+
+[Online demo](https://huggingface.co/spaces/depth-anything/Depth-Anything-V2).
+
+## Model description
+
+Depth Anything V2 leverages the [DPT](https://huggingface.co/docs/transformers/model_doc/dpt) architecture with a [DINOv2](https://huggingface.co/docs/transformers/model_doc/dinov2) backbone.
+
+The model is trained on ~600K synthetic labeled images and ~62 million real unlabeled images, obtaining state-of-the-art results for both relative and absolute depth estimation.
+
+<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/depth_anything_overview.jpg"
+alt="drawing" width="600"/>
+
+<small> Depth Anything overview. Taken from the <a href="https://arxiv.org/abs/2401.10891">original paper</a>.</small>
+
+## Intended uses & limitations
+
+You can use the raw model for tasks like zero-shot depth estimation. See the [model hub](https://huggingface.co/models?search=depth-anything) to look for
+other versions on a task that interests you.
 
-## Usage
+### How to use
 
-Download the [model](https://huggingface.co/depth-anything/Depth-Anything-V2-Large/resolve/main/depth_anything_v2_vitl.pth?download=true) first and put it under the `checkpoints` directory.
+Here is how to use this model to perform zero-shot depth estimation:
 
 ```python
-import cv2
+from transformers import pipeline
+from PIL import Image
+import requests
+
+# load pipe
+pipe = pipeline(task="depth-estimation", model="depth-anything/Depth-Anything-V2-Large-hf")
+
+# load image
+url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
+image = Image.open(requests.get(url, stream=True).raw)
+
+# inference
+depth = pipe(image)["depth"]
+```
+
+Alternatively, you can use the model and processor classes:
+
+```python
+from transformers import AutoImageProcessor, AutoModelForDepthEstimation
 import torch
+import numpy as np
+from PIL import Image
+import requests
+
+url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+image = Image.open(requests.get(url, stream=True).raw)
+
+image_processor = AutoImageProcessor.from_pretrained("depth-anything/Depth-Anything-V2-Large-hf")
+model = AutoModelForDepthEstimation.from_pretrained("depth-anything/Depth-Anything-V2-Large-hf")
 
-from depth_anything_v2.dpt import DepthAnythingV2
+# prepare image for the model
+inputs = image_processor(images=image, return_tensors="pt")
 
-model = DepthAnythingV2(encoder='vitl', features=256, out_channels=[256, 512, 1024, 1024])
-model.load_state_dict(torch.load('checkpoints/depth_anything_v2_vitl.pth', map_location='cpu'))
-model.eval()
+with torch.no_grad():
+    outputs = model(**inputs)
+    predicted_depth = outputs.predicted_depth
 
-raw_img = cv2.imread('your/image/path')
-depth = model.infer_image(raw_img) # HxW raw depth map
+# interpolate to original size
+prediction = torch.nn.functional.interpolate(
+    predicted_depth.unsqueeze(1),
+    size=image.size[::-1],
+    mode="bicubic",
+    align_corners=False,
+)
 ```
 
-## Citation
+For more code examples, please refer to the [documentation](https://huggingface.co/transformers/main/model_doc/depth_anything.html#).
 
-If you find this project useful, please consider citing:
+
+### Citation
 
 ```bibtex
-@article{depth_anything_v2,
-  title={Depth Anything V2},
-  author={Yang, Lihe and Kang, Bingyi and Huang, Zilong and Zhao, Zhen and Xu, Xiaogang and Feng, Jiashi and Zhao, Hengshuang},
-  journal={arXiv:2406.09414},
-  year={2024}
+@misc{yang2024depth,
+      title={Depth Anything V2}, 
+      author={Lihe Yang and Bingyi Kang and Zilong Huang and Zhen Zhao and Xiaogang Xu and Jiashi Feng and Hengshuang Zhao},
+      year={2024},
+      eprint={2406.09414},
+      archivePrefix={arXiv},
+      primaryClass={id='cs.CV' full_name='Computer Vision and Pattern Recognition' is_active=True alt_name=None in_archive='cs' is_general=False description='Covers image processing, computer vision, pattern recognition, and scene understanding. Roughly includes material in ACM Subject Classes I.2.10, I.4, and I.5.'}
 }
-
-@inproceedings{depth_anything_v1,
-  title={Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data}, 
-  author={Yang, Lihe and Kang, Bingyi and Huang, Zilong and Xu, Xiaogang and Feng, Jiashi and Zhao, Hengshuang},
-  booktitle={CVPR},
-  year={2024}
-}
+```

config.json

@@ -0,0 +1,81 @@
+{
+  "_commit_hash": null,
+  "architectures": [
+    "DepthAnythingForDepthEstimation"
+  ],
+  "backbone": null,
+  "backbone_config": {
+    "architectures": [
+      "Dinov2Model"
+    ],
+    "hidden_size": 1024,
+    "image_size": 518,
+    "model_type": "dinov2",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 24,
+    "out_features": [
+      "stage5",
+      "stage12",
+      "stage18",
+      "stage24"
+    ],
+    "out_indices": [
+      5,
+      12,
+      18,
+      24
+    ],
+    "patch_size": 14,
+    "reshape_hidden_states": false,
+    "stage_names": [
+      "stem",
+      "stage1",
+      "stage2",
+      "stage3",
+      "stage4",
+      "stage5",
+      "stage6",
+      "stage7",
+      "stage8",
+      "stage9",
+      "stage10",
+      "stage11",
+      "stage12",
+      "stage13",
+      "stage14",
+      "stage15",
+      "stage16",
+      "stage17",
+      "stage18",
+      "stage19",
+      "stage20",
+      "stage21",
+      "stage22",
+      "stage23",
+      "stage24"
+    ],
+    "torch_dtype": "float32"
+  },
+  "fusion_hidden_size": 256,
+  "head_hidden_size": 32,
+  "head_in_index": -1,
+  "initializer_range": 0.02,
+  "model_type": "depth_anything",
+  "neck_hidden_sizes": [
+    256,
+    512,
+    1024,
+    1024
+  ],
+  "patch_size": 14,
+  "reassemble_factors": [
+    4,
+    2,
+    1,
+    0.5
+  ],
+  "reassemble_hidden_size": 1024,
+  "torch_dtype": "float32",
+  "transformers_version": null,
+  "use_pretrained_backbone": false
+}

depth_anything_v2_vitl.pth → model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7ea19fa0ed99244e67b624c72b8580b7e9553043245905be58796a608eb9345
-size 1341395338
+oid sha256:4e01e34ed5549b529b70b92d53226bc370f03041977b390d3dde45d47f516cf9
+size 1341322868

preprocessor_config.json

@@ -0,0 +1,44 @@
+{
+  "_valid_processor_keys": [
+    "images",
+    "do_resize",
+    "size",
+    "keep_aspect_ratio",
+    "ensure_multiple_of",
+    "resample",
+    "do_rescale",
+    "rescale_factor",
+    "do_normalize",
+    "image_mean",
+    "image_std",
+    "do_pad",
+    "size_divisor",
+    "return_tensors",
+    "data_format",
+    "input_data_format"
+  ],
+  "do_normalize": true,
+  "do_pad": false,
+  "do_rescale": true,
+  "do_resize": true,
+  "ensure_multiple_of": 14,
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "DPTImageProcessor",
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "keep_aspect_ratio": true,
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 518,
+    "width": 518
+  },
+  "size_divisor": null
+}