Add Github repository content

DA-2K.md

@@ -0,0 +1,51 @@
+# DA-2K Evaluation Benchmark
+
+## Introduction
+
+![DA-2K](assets/DA-2K.png)
+
+DA-2K is proposed in [Depth Anything V2](https://depth-anything-v2.github.io) to evaluate the relative depth estimation capability. It encompasses eight representative scenarios of `indoor`, `outdoor`, `non_real`, `transparent_reflective`, `adverse_style`, `aerial`, `underwater`, and `object`. It consists of 1K diverse high-quality images and 2K precise pair-wise relative depth annotations.
+
+Please refer to our [paper](https://arxiv.org/abs/2406.09414) for details in constructing this benchmark.
+
+
+## Usage
+
+Please first [download the benchmark](https://huggingface.co/datasets/depth-anything/DA-2K/tree/main).
+
+All annotations are stored in `annotations.json`. The annotation file is a JSON object where each key is the path to an image file, and the value is a list of annotations associated with that image. Each annotation describes two points and identifies which point is closer to the camera. The structure is detailed below:
+
+```
+{
+  "image_path": [
+    {
+      "point1": [h1, w1], # (vertical position, horizontal position)
+      "point2": [h2, w2], # (vertical position, horizontal position)
+      "closer_point": "point1" # we always set "point1" as the closer one
+    },
+    ...
+  ],
+  ...
+}
+```
+
+To visualize the annotations:
+```bash
+python visualize.py [--scene-type <type>]
+```
+
+**Options**
+- `--scene-type <type>` (optional): Specify the scene type (`indoor`, `outdoor`, `non_real`, `transparent_reflective`, `adverse_style`, `aerial`, `underwater`, and `object`). Skip this argument or set <type> as `""` to include all scene types.
+
+## Citation
+
+If you find this benchmark useful, please consider citing:
+
+```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}
+}
+```

README.md

@@ -1,13 +1,136 @@
----
-title: Depth Anything V2
-emoji: 🌖
-colorFrom: red
-colorTo: indigo
-sdk: gradio
-sdk_version: 4.36.0
-app_file: app.py
-pinned: false
-license: bsd-2-clause
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+<div align="center">
+<h1>Depth Anything V2</h1>
+
+[**Lihe Yang**](https://liheyoung.github.io/)<sup>1</sup> · [**Bingyi Kang**](https://bingykang.github.io/)<sup>2&dagger;</sup> · [**Zilong Huang**](http://speedinghzl.github.io/)<sup>2</sup>
+<br>
+[**Zhen Zhao**](http://zhaozhen.me/) · [**Xiaogang Xu**](https://xiaogang00.github.io/) · [**Jiashi Feng**](https://sites.google.com/site/jshfeng/)<sup>2</sup> · [**Hengshuang Zhao**](https://hszhao.github.io/)<sup>1*</sup>
+
+<sup>1</sup>HKU&emsp;&emsp;&emsp;<sup>2</sup>TikTok
+<br>
+&dagger;project lead&emsp;*corresponding author
+
+<a href="https://arxiv.org/abs/2406.09414"><img src='https://img.shields.io/badge/arXiv-Depth Anything V2-red' alt='Paper PDF'></a>
+<a href='https://depth-anything-v2.github.io'><img src='https://img.shields.io/badge/Project_Page-Depth Anything V2-green' alt='Project Page'></a>
+<a href='https://huggingface.co/spaces/depth-anything/Depth-Anything-V2'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue'></a>
+<a href='https://huggingface.co/datasets/depth-anything/DA-2K'><img src='https://img.shields.io/badge/Benchmark-DA--2K-yellow' alt='Benchmark'></a>
+</div>
+
+This work presents Depth Anything V2. It significantly outperforms [V1](https://github.com/LiheYoung/Depth-Anything) in fine-grained details and robustness. Compared with SD-based models, it enjoys faster inference speed, fewer parameters, and higher depth accuracy.
+
+![teaser](assets/teaser.png)
+
+## News
+
+- **2024-06-14:** Paper, project page, code, models, demo, and benchmark are all released.
+
+
+## Pre-trained Models
+
+We provide **four models** of varying scales for robust relative depth estimation:
+
+| Model | Params | Checkpoint |
+|:-|-:|:-:|
+| Depth-Anything-V2-Small | 24.8M | [Download](https://huggingface.co/depth-anything/Depth-Anything-V2-Small/resolve/main/depth_anything_v2_vits.pth?download=true) |
+| Depth-Anything-V2-Base | 97.5M | [Download](https://huggingface.co/depth-anything/Depth-Anything-V2-Base/resolve/main/depth_anything_v2_vitb.pth?download=true) |
+| Depth-Anything-V2-Large | 335.3M | [Download](https://huggingface.co/depth-anything/Depth-Anything-V2-Large/resolve/main/depth_anything_v2_vitl.pth?download=true) |
+| Depth-Anything-V2-Giant | 1.3B | Coming soon |
+
+
+### Code snippet to use our models
+```python
+import cv2
+import torch
+
+from depth_anything_v2.dpt import DepthAnythingV2
+
+# take depth-anything-v2-large as an example
+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()
+
+raw_img = cv2.imread('your/image/path')
+depth = model.infer_image(raw_img) # HxW raw depth map
+```
+
+## Usage
+
+### Installation
+
+```bash
+git clone https://github.com/DepthAnything/Depth-Anything-V2
+cd Depth-Anything-V2
+pip install -r requirements.txt
+```
+
+### Running
+
+```bash
+python run.py --encoder <vits | vitb | vitl | vitg> --img-path <path> --outdir <outdir> [--input-size <size>] [--pred-only] [--grayscale]
+```
+Options:
+- `--img-path`: You can either 1) point it to an image directory storing all interested images, 2) point it to a single image, or 3) point it to a text file storing all image paths.
+- `--input-size` (optional): By default, we use input size `518` for model inference. **You can increase the size for even more fine-grained results.**
+- `--pred-only` (optional): Only save the predicted depth map, without raw image.
+- `--grayscale` (optional): Save the grayscale depth map, without applying color palette.
+
+For example:
+```bash
+python run.py --encoder vitl --img-path assets/examples --outdir depth_vis
+```
+
+**If you want to use Depth Anything V2 on videos:**
+
+```bash
+python run_video.py --encoder vitl --video-path assets/examples_video --outdir video_depth_vis
+```
+
+*Please note that our larger model has better temporal consistency on videos.*
+
+
+### Gradio demo
+
+To use our gradio demo locally:
+
+```bash
+python app.py
+```
+
+You can also try our [online demo](https://huggingface.co/spaces/Depth-Anything/Depth-Anything-V2).
+
+**Note:** Compared to V1, we have made a minor modification to the DINOv2-DPT architecture (originating from this [issue](https://github.com/LiheYoung/Depth-Anything/issues/81)). In V1, we *unintentionally* used features from the last four layers of DINOv2 for decoding. In V2, we use [intermediate features](https://github.com/DepthAnything/Depth-Anything-V2/blob/2cbc36a8ce2cec41d38ee51153f112e87c8e42d8/depth_anything_v2/dpt.py#L164-L169) instead. Although this modification did not improve details or accuracy, we decided to follow this common practice.
+
+
+
+## Fine-tuned to Metric Depth Estimation
+
+Please refer to [metric depth estimation](./metric_depth).
+
+
+## DA-2K Evaluation Benchmark
+
+Please refer to [DA-2K benchmark](./DA-2K.md).
+
+## LICENSE
+
+Depth-Anything-V2-Small model is under the Apache-2.0 license. Depth-Anything-V2-Base/Large/Giant models are under the CC-BY-NC-4.0 license.
+
+
+## Citation
+
+If you find this project useful, please consider citing:
+
+```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}
+}
+
+@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}
+}
+```

run.py

@@ -0,0 +1,74 @@
+import argparse
+import cv2
+import glob
+import matplotlib
+import numpy as np
+import os
+import torch
+
+from depth_anything_v2.dpt import DepthAnythingV2
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description='Depth Anything V2')
+    
+    parser.add_argument('--img-path', type=str)
+    parser.add_argument('--input-size', type=int, default=518)
+    parser.add_argument('--outdir', type=str, default='./vis_depth')
+    
+    parser.add_argument('--encoder', type=str, default='vitl', choices=['vits', 'vitb', 'vitl', 'vitg'])
+    
+    parser.add_argument('--pred-only', dest='pred_only', action='store_true', help='only display the prediction')
+    parser.add_argument('--grayscale', dest='grayscale', action='store_true', help='do not apply colorful palette')
+    
+    args = parser.parse_args()
+    
+    DEVICE = 'cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu'
+    
+    # we are undergoing company review procedures to release Depth-Anything-Giant checkpoint
+    model_configs = {
+        'vits': {'encoder': 'vits', 'features': 64, 'out_channels': [48, 96, 192, 384]},
+        'vitb': {'encoder': 'vitb', 'features': 128, 'out_channels': [96, 192, 384, 768]},
+        'vitl': {'encoder': 'vitl', 'features': 256, 'out_channels': [256, 512, 1024, 1024]},
+        'vitg': {'encoder': 'vitg', 'features': 384, 'out_channels': [1536, 1536, 1536, 1536]}
+    }
+    
+    depth_anything = DepthAnythingV2(**model_configs[args.encoder])
+    depth_anything.load_state_dict(torch.load(f'checkpoints/depth_anything_v2_{args.encoder}.pth', map_location='cpu'))
+    depth_anything = depth_anything.to(DEVICE).eval()
+    
+    if os.path.isfile(args.img_path):
+        if args.img_path.endswith('txt'):
+            with open(args.img_path, 'r') as f:
+                filenames = f.read().splitlines()
+        else:
+            filenames = [args.img_path]
+    else:
+        filenames = glob.glob(os.path.join(args.img_path, '**/*'), recursive=True)
+    
+    os.makedirs(args.outdir, exist_ok=True)
+    
+    cmap = matplotlib.colormaps.get_cmap('Spectral_r')
+    
+    for k, filename in enumerate(filenames):
+        print(f'Progress {k+1}/{len(filenames)}: {filename}')
+        
+        raw_image = cv2.imread(filename)
+        
+        depth = depth_anything.infer_image(raw_image, args.input_size)
+        
+        depth = (depth - depth.min()) / (depth.max() - depth.min()) * 255.0
+        depth = depth.astype(np.uint8)
+        
+        if args.grayscale:
+            depth = np.repeat(depth[..., np.newaxis], 3, axis=-1)
+        else:
+            depth = (cmap(depth)[:, :, :3] * 255)[:, :, ::-1].astype(np.uint8)
+        
+        if args.pred_only:
+            cv2.imwrite(os.path.join(args.outdir, os.path.splitext(os.path.basename(filename))[0] + '.png'), depth)
+        else:
+            split_region = np.ones((raw_image.shape[0], 50, 3), dtype=np.uint8) * 255
+            combined_result = cv2.hconcat([raw_image, split_region, depth])
+            
+            cv2.imwrite(os.path.join(args.outdir, os.path.splitext(os.path.basename(filename))[0] + '.png'), combined_result)

run_video.py

@@ -0,0 +1,89 @@
+import argparse
+import cv2
+import glob
+import matplotlib
+import numpy as np
+import os
+import torch
+
+from depth_anything_v2.dpt import DepthAnythingV2
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description='Depth Anything V2')
+    
+    parser.add_argument('--video-path', type=str)
+    parser.add_argument('--input-size', type=int, default=518)
+    parser.add_argument('--outdir', type=str, default='./vis_video_depth')
+    
+    parser.add_argument('--encoder', type=str, default='vitl', choices=['vits', 'vitb', 'vitl', 'vitg'])
+    
+    parser.add_argument('--pred-only', dest='pred_only', action='store_true', help='only display the prediction')
+    parser.add_argument('--grayscale', dest='grayscale', action='store_true', help='do not apply colorful palette')
+    
+    args = parser.parse_args()
+    
+    DEVICE = 'cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu'
+    
+    # 'we are undergoing company review procedures to release Depth-Anything-Giant checkpoint
+    model_configs = {
+        'vits': {'encoder': 'vits', 'features': 64, 'out_channels': [48, 96, 192, 384]},
+        'vitb': {'encoder': 'vitb', 'features': 128, 'out_channels': [96, 192, 384, 768]},
+        'vitl': {'encoder': 'vitl', 'features': 256, 'out_channels': [256, 512, 1024, 1024]},
+        'vitg': {'encoder': 'vitg', 'features': 384, 'out_channels': [1536, 1536, 1536, 1536]}
+    }
+    
+    depth_anything = DepthAnythingV2(**model_configs[args.encoder])
+    depth_anything.load_state_dict(torch.load(f'checkpoints/depth_anything_v2_{args.encoder}.pth', map_location='cpu'))
+    depth_anything = depth_anything.to(DEVICE).eval()
+    
+    if os.path.isfile(args.video_path):
+        if args.video_path.endswith('txt'):
+            with open(args.video_path, 'r') as f:
+                lines = f.read().splitlines()
+        else:
+            filenames = [args.video_path]
+    else:
+        filenames = glob.glob(os.path.join(args.video_path, '**/*'), recursive=True)
+    
+    os.makedirs(args.outdir, exist_ok=True)
+    
+    margin_width = 50
+    cmap = matplotlib.colormaps.get_cmap('Spectral_r')
+    
+    for k, filename in enumerate(filenames):
+        print(f'Progress {k+1}/{len(filenames)}: {filename}')
+        
+        raw_video = cv2.VideoCapture(filename)
+        frame_width, frame_height = int(raw_video.get(cv2.CAP_PROP_FRAME_WIDTH)), int(raw_video.get(cv2.CAP_PROP_FRAME_HEIGHT))
+        frame_rate = int(raw_video.get(cv2.CAP_PROP_FPS))
+        output_width = frame_width * 2 + margin_width
+        
+        output_path = os.path.join(args.outdir, os.path.splitext(os.path.basename(filename))[0] + '.mp4')
+        out = cv2.VideoWriter(output_path, cv2.VideoWriter_fourcc(*"mp4v"), frame_rate, (output_width, frame_height))
+        
+        while raw_video.isOpened():
+            ret, raw_frame = raw_video.read()
+            if not ret:
+                break
+            
+            depth = depth_anything.infer_image(raw_frame, args.input_size)
+            
+            depth = (depth - depth.min()) / (depth.max() - depth.min()) * 255.0
+            depth = depth.astype(np.uint8)
+            
+            if args.grayscale:
+                depth = np.repeat(depth[..., np.newaxis], 3, axis=-1)
+            else:
+                depth = (cmap(depth)[:, :, :3] * 255)[:, :, ::-1].astype(np.uint8)
+            
+            if args.pred_only:
+                out.write(depth)
+            else:
+                split_region = np.ones((frame_height, margin_width, 3), dtype=np.uint8) * 255
+                combined_frame = cv2.hconcat([raw_frame, split_region, depth])
+                
+                out.write(combined_frame)
+        
+        raw_video.release()
+        out.release()