update annotator

.gitignore

@@ -1,4 +1,5 @@
 annotator/ckpts/**
+annotator/annotator_ckpts.tar.gz
 result/**
 trash/**
 data/** 

README.md

@@ -24,11 +24,37 @@ pip install -r requirements.txt
 
 </details>
 
-You may download all data and checkpoints using the following bash command
+You may download all the base model checkpoints using the following bash command
 ```bash
+## download sd 1.5, controlnet depth/pose v10/v11
 bash download_all.sh
 ```
 
+Prepare ControlNet annotator weights (e.g., DW-Pose, depth_zoe, depth_midas, OpenPose)
+
+```bash
+## Switch to the ckpts directory
+cd ckpts
+
+Method 1: Download individual models
+Download the DW-Pose models (dw-ll_ucoco_384.onnx and yolo_l.onnx), as we found them to be more robust than OpenPose. (Note: Other models, such as depth_zoe, depth_midas, and OpenPose, can be automatically downloaded from HuggingFace.)
+Available from:
+  - [Baidu](https://pan.baidu.com/s/1nuBjw-KKSxD_BkpmwXUJiw?pwd=28d7)
+  - [Google](https://drive.google.com/file/d/12L8E2oAgZy4VACGSK9RaZBZrfgx7VTA2/view?usp=sharing)
+Download the detection model (yolox_l.onnx) from:
+  - [Baidu](https://pan.baidu.com/s/1fpfIVpv5ypo4c1bUlzkMYQ?pwd=mjdn)
+  - [Google](https://drive.google.com/file/d/1w9pXC8tT0p9ndMN-CArp1__b2GbzewWI/view?usp=sharing)
+Then place both files into ./annotator/ckpts
+
+Method 2: Download all annotator checkpoints in one package
+# Note: This package includes all the required annotator models, such as DW-Pose, depth_zoe, depth_midas, and OpenPose. Requires approximately 4GB of storage space.
+If you cannot access HuggingFace, you can download all the annotator checkpoints from:
+  - [BaiduYun](https://pan.baidu.com/s/1sgBFLFkdTCDTn4oqHjGb9A?pwd=pdm5)
+  - [Google](https://drive.google.com/file/d/1qOsmWshnFMMr8x1HteaTViTSQLh_4rle/view?usp=drive_link)
+Then extract them into ./annotator/ckpts
+
+```
+
 ## 🔛 Prepare all the data
 
 ```