Upload

.gitattributes

@@ -33,3 +33,19 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoints/modelscope_ocr/damo/cv_convnextTiny_ocr-recognition-general_damo/resources/ConvTransformer-Pipeline.jpg filter=lfs diff=lfs merge=lfs -text
+Imgs/failure_case.jpg filter=lfs diff=lfs merge=lfs -text
+Imgs/marconet_vs_marconetplus.jpg filter=lfs diff=lfs merge=lfs -text
+Imgs/pipeline.png filter=lfs diff=lfs merge=lfs -text
+Imgs/prior.gif filter=lfs diff=lfs merge=lfs -text
+Imgs/text_line_sr.jpg filter=lfs diff=lfs merge=lfs -text
+Imgs/w-interpolation.gif filter=lfs diff=lfs merge=lfs -text
+Imgs/whole_1.jpg filter=lfs diff=lfs merge=lfs -text
+Imgs/whole_2.jpg filter=lfs diff=lfs merge=lfs -text
+Imgs/whole_3.jpg filter=lfs diff=lfs merge=lfs -text
+Imgs/whole_4.jpg filter=lfs diff=lfs merge=lfs -text
+Testsets/LR_Whole/ip11pro_output_testSR_2101225_xwm_renew_12.JPG filter=lfs diff=lfs merge=lfs -text
+Testsets/LR_Whole/ip11pro_output_testSR_2101225_xwm_renew_16.JPG filter=lfs diff=lfs merge=lfs -text
+Testsets/LR_Whole/ip11pro_output_testSR_2101225_xwm_renew_8.JPG filter=lfs diff=lfs merge=lfs -text
+Testsets/LR_Whole/ip11pro_output_textSR_211206_renew_30.JPG filter=lfs diff=lfs merge=lfs -text
+Testsets/LR_Whole/ip11pro_output_textSR_211206_renew_62.JPG filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,9 @@
+S-Lab License 1.0  
+
+Copyright 2025 S-Lab
+Redistribution and use for non-commercial purpose in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
+1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
+2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
+3. Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+4. In the event that redistribution and/or use for commercial purpose in source or binary forms, with or without modification is required, please contact the contributor(s) of the work.

README.md

@@ -1,14 +1,147 @@
----
-title: Marconetplusplus
-emoji: 🦀
-colorFrom: yellow
-colorTo: yellow
-sdk: gradio
-sdk_version: 5.42.0
-app_file: app.py
-pinned: false
-license: cc-by-nc-4.0
-short_description: Chinese Text Image Super-Resolution
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+<div align="center">
+  
+  ## [Enhanced Generative Structure Prior for Chinese Text Image Super-Resolution](https://arxiv.org/pdf/2508.07537)
+
+[Xiaoming Li](https://csxmli2016.github.io/), [Wangmeng Zuo](https://scholar.google.com/citations?hl=en&user=rUOpCEYAAAAJ&view_op=list_works), [Chen Change Loy](https://www.mmlab-ntu.com/person/ccloy/)
+
+S-Lab, Nanyang Technological University
+
+</div>
+
+
+### 👷 The whole framework:
+<div align="center">
+<img src="./Imgs/pipeline.png" width="800px">
+</div>
+
+
+
+<!-- 
+<p align="justify">Faithful text image super-resolution (SR) is challenging because each character has a unique structure and usually exhibits diverse font styles and
+layouts. While existing methods primarily focus on English text, less attention has been paid to more complex scripts like Chinese. In this paper, we introduce a high-quality text image SR framework designed to restore the precise strokes of low-resolution (LR) Chinese characters. Unlike methods that rely on character recognition priors to regularize the SR task, we propose a novel structure prior that offers structure-level guidance to enhance visual quality. Our framework incorporates this structure prior within a StyleGAN model, leveraging its generative capabilities for restoration. To maintain the integrity of character structures while accommodating various font styles and layouts, we implement a codebook-based mechanism that restricts the generative space of StyleGAN. Each code in the codebook represents the structure of a specific character, while the vector $w$ in StyleGAN controls the character's style, including typeface, orientation, and location. Through the collaborative interaction between the codebook and style, we generate a high-resolution structure prior that aligns with LR characters both spatially and structurally. Experiments demonstrate that this structure prior provides robust, character-specific guidance, enabling the accurate restoration of clear strokes in degraded characters, even for real-world LR Chinese text with irregular layouts.  </p>
+-->
+
+### 👷 Character Structure Prior Pretraining:
+<div align="center">
+<img src="./Imgs/prior.gif" width="800px">
+</div>
+
+
+## 🔔 MARCONet 🆚 MARCONet++
+> - MARCONet is designed for **regular character layout** only. See details of [MARCONet](https://github.com/csxmli2016/MARCONet).
+> - MARCONet++ has more accurate alignment between character structural prior (green structure) and the degraded image.
+<div align="center">
+<img src="./Imgs/marconet_vs_marconetplus.jpg" width="800px">
+</div>
+
+
+
+## 📋 TODO
+- [x] Release the inference code and model.
+- [ ] Release the training code (no plans to release for now). 
+
+
+## 🚶 Getting Started
+
+```
+git clone https://github.com/csxmli2016/MARCONetPlusPlus
+cd MARCONetPlusPlus
+conda create -n mplus python=3.8 -y
+conda activate mplus
+pip install -r requirements.txt
+```
+
+## 🚶 Inference
+Download the pre-trained models
+```
+python utils/download_github.py
+```
+
+and run for restoring **text lines:**
+```
+CUDA_VISIBLE_DEVICES=0 python test_marconetplus.py -i ./Testsets/LR_TextLines -a -s
+```
+or run for restoring **the whole text image:**
+```
+CUDA_VISIBLE_DEVICES=0 python test_marconetplus.py -i ./Testsets/LR_Whole -b -s -f 2
+```
+
+```
+# Parameters:
+-i: --input_path, default: ./Testsets/LR_TextLines or ./Testsets/LR_TextWhole
+-o: --output_path, default: None will automatically make the saving dir with the format of '[LR path]_TIME_MARCONetPlus'
+-a: --aligned, if the input is text lines, use -a; otherwise, the input is the whole text image and needs text line detection, do not use -a
+-b: --bg_sr, when restoring the whole text images, use -b to restore the background region with BSRGAN. Without -b, background will keep the same as input
+-f: --factor_scale, default: 2. When restoring the whole text images, use -f to define the scale factor of output
+-s: --save_text, if you want to see the details of prior alignment, predicted characters, and locations, use -s 
+```
+
+
+
+## 🏃 Restoring Real-world Chinese Text Images
+> - We use [BSRGAN](https://github.com/cszn/BSRGAN) to restore the background region.
+> - The parameters are tested on an NVIDIA A100 GPU (40G).
+> - ⚠️ If the inference speed is slow, this is caused by the large size of the input text image or the large factor_scale. You can resize it based on your needs.
+
+[<img src="Imgs/whole_1.jpg" height="270px"/>](https://imgsli.com/NDA2MDUw) [<img src="Imgs/whole_2.jpg" height="270px"/>](https://imgsli.com/NDA2MDYw) 
+
+[<img src="Imgs/whole_3.jpg" height="540px"/>](https://imgsli.com/NDA2MTE0) [<img src="Imgs/whole_4.jpg" height="540px" width="418px"/>](https://imgsli.com/NDA2MDYy) 
+
+
+## 🏃 Restoring detected text line
+
+<img src="Imgs/text_line_sr.jpg" width="800px"/>
+
+
+<details><summary><h2>🏃 Style w interpolation from three characters with different styles</h2></summary>
+  <img src="./Imgs/w-interpolation.gif" width="400px">
+</details>
+
+
+
+
+## ‼️ Failure Case
+Despite its high-fidelity performance, MARCONet++ still struggles in some real-world scenarios as it highly relies on:
+
+- Real world character **Recognition** on complex degraded text images
+- Real world character **Detection** on complex degraded text images
+- Text line detection and segmentation
+- Domain gap between our synthetic and real-world text images
+
+
+<img src="./Imgs/failure_case.jpg"  width="800px">
+
+> 🍒 Restoring complex character with high fidelity under such conditions has significant challenges.
+We have also explored various approaches, such as training OCR models with Transformers and using YOLO or Transformer-based methods for character detection, but these methods generally encounter the same issues. 
+We encourage any potential collaborations to jointly tackle this challenge and advance robust, high-fidelity text restoration.
+
+
+## 📎 RealCE-1K benchmark
+To quantitatively evaluate on real-world Chinese text line images, we curate a benchmark by filtering the [RealCE](https://github.com/mjq11302010044/Real-CE) test set to exclude images containing multiple text lines or inaccurate annotations, thereby constructing a Chinese text SR benchmark (see Section IV.B of our paper). You can download the RealCE-1K benchmark from [here](https://github.com/csxmli2016/MARCONetPlusPlus/releases/download/v1/RealCE-1K.zip).
+
+## 🍺 Acknowledgement
+This project is built based on the excellent [KAIR](https://github.com/cszn/KAIR) and [RealCE](https://github.com/mjq11302010044/Real-CE).
+
+## ©️ License
+This project is licensed under <a rel="license" href="https://github.com/csxmli2016/MARCONetPlusPlus/blob/main/LICENSE">NTU S-Lab License 1.0</a>. Redistribution and use should follow this license.
+
+
+## 🍻 Citation
+```
+@article{li2025marconetplus,
+  author = {Li, Xiaoming and Zuo, Wangmeng and Loy, Chen Change},
+  title = {Enhanced Generative Structure Prior for Chinese Text Image Super-Resolution},
+  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
+  year = {2025}
+}
+
+@InProceedings{li2023marconet,
+  author = {Li, Xiaoming and Zuo, Wangmeng and Loy, Chen Change},
+  title = {Learning Generative Structure Prior for Blind Text Image Super-resolution},
+  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
+  year = {2023}
+}
+```
+
+

app.py

@@ -0,0 +1,117 @@
+import torch
+import cv2
+import numpy as np
+import os
+import os.path as osp
+import time
+import gradio as gr
+
+os.environ["GRADIO_TEMP_DIR"] = "./gradio_tmp"
+
+from models.TextEnhancement import MARCONetPlus
+from utils.utils_image import imread_uint, uint2tensor4, tensor2uint
+from networks.rrdbnet2_arch import RRDBNet as BSRGAN
+
+# Initialize device
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+
+# Background restoration model (lazy loading)
+BGModel = None
+def load_bg_model():
+    """Load BSRGAN model for background super-resolution"""
+    global BGModel
+    if BGModel is None:
+        BGModel = BSRGAN(in_nc=3, out_nc=3, nf=64, nb=23, gc=32, sf=2)
+        model_old = torch.load('./checkpoints/bsrgan_bg.pth', map_location=device)
+        state_dict = BGModel.state_dict()
+        for ((key, param), (key2, _)) in zip(model_old.items(), state_dict.items()):
+            state_dict[key2] = param
+        BGModel.load_state_dict(state_dict, strict=True)
+        BGModel.eval()
+        for k, v in BGModel.named_parameters():
+            v.requires_grad = False
+        BGModel = BGModel.to(device)
+
+# Text restoration model
+TextModel = MARCONetPlus(
+    './checkpoints/net_w_encoder_860000.pth',
+    './checkpoints/net_prior_860000.pth',
+    './checkpoints/net_sr_860000.pth',
+    './checkpoints/yolo11m_short_character.pt',
+    device=device
+)
+
+def gradio_inference(input_img, aligned=False, bg_sr=False, scale_factor=2):
+    """Run MARCONetPlus inference with optional background SR"""
+    if input_img is None:
+        return None
+
+    # Convert input image (PIL) to OpenCV format
+    img_L = cv2.cvtColor(np.array(input_img), cv2.COLOR_RGB2BGR)
+    height_L, width_L = img_L.shape[:2]
+
+    # Background super-resolution
+    if not aligned and bg_sr:
+        load_bg_model()
+        img_E = cv2.resize(img_L, (int(width_L//8*8), int(height_L//8*8)), interpolation=cv2.INTER_AREA)
+        img_E = uint2tensor4(img_E).to(device)
+        with torch.no_grad():
+            try:
+                img_E = BGModel(img_E)
+            except:
+                torch.cuda.empty_cache()
+                max_size = 1536
+                scale = min(max_size / width_L, max_size / height_L, 1.0)
+                new_width = int(width_L * scale)
+                new_height = int(height_L * scale)
+                img_E = cv2.resize(img_L, (new_width//8*8, new_height//8*8), interpolation=cv2.INTER_AREA)
+                img_E = uint2tensor4(img_E).to(device)
+                img_E = BGModel(img_E)
+        img_E = tensor2uint(img_E)
+    else:
+        img_E = img_L
+
+    # Resize background
+    width_S = width_L * scale_factor
+    height_S = height_L * scale_factor
+    img_E = cv2.resize(img_E, (width_S, height_S), interpolation=cv2.INTER_AREA)
+
+    # Text restoration
+    SQ, ori_texts, en_texts, debug_texts, pred_texts = TextModel.handle_texts(
+        img=img_L, bg=img_E, sf=scale_factor, is_aligned=aligned
+    )
+
+    if SQ is None:
+        return None
+
+    if not aligned:
+        SQ = cv2.resize(SQ.astype(np.float32), (width_S, height_S), interpolation=cv2.INTER_AREA)
+        out_img = SQ[:, :, ::-1].astype(np.uint8)
+    else:
+        out_img = en_texts[0][:, :, ::-1].astype(np.uint8)
+
+    return out_img
+
+# Gradio UI
+with gr.Blocks() as demo:
+    gr.Markdown("# MARCONetPlus Text Image Restoration")
+
+    with gr.Row():
+        input_img = gr.Image(type="pil", label="Input Image")
+        output_img = gr.Image(type="numpy", label="Restored Output")
+
+    with gr.Row():
+        aligned = gr.Checkbox(label="Aligned (cropped text regions)", value=False)
+        bg_sr = gr.Checkbox(label="Background SR (BSRGAN)", value=False)
+        scale_factor = gr.Slider(1, 4, value=2, step=1, label="Scale Factor")
+
+    run_btn = gr.Button("Run Inference")
+
+    run_btn.click(
+        fn=gradio_inference,
+        inputs=[input_img, aligned, bg_sr, scale_factor],
+        outputs=[output_img]
+    )
+
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0", server_port=7121)

checkpoints/bsrgan_bg.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82df5fa543c716933bb26592519848c74ada4c86f101324ecb8e9af863c61793
+size 66894171

checkpoints/modelscope_ocr/damo/cv_convnextTiny_ocr-recognition-general_damo/.mv

@@ -0,0 +1 @@
+Revision:v2.4.0,CreatedAt:1695105029

checkpoints/modelscope_ocr/damo/cv_convnextTiny_ocr-recognition-general_damo/README.md

@@ -0,0 +1,245 @@
+---
+backbone:
+- convNext-Tiny
+integrating: True
+domain:
+- cv
+frameworks:
+- pytorch
+language:
+- en
+- ch
+license: Apache License 2.0
+metrics:
+- Line Accuracy
+finetune-support: True
+tags:
+- OCR
+- Alibaba
+- 文字识别
+- 读光
+tasks:
+- ocr-recognition
+
+studios:
+- damo/cv_ocr-text-spotting
+
+datasets:
+  test:
+  - damo/WebText_Dataset
+
+widgets:
+  - task: ocr-recognition
+    inputs:
+      - type: image
+    examples:
+      - name: 1
+        inputs:
+          - name: image
+            data: http://duguang-labelling.oss-cn-shanghai.aliyuncs.com/mass_img_tmp_20220922/ocr_recognition.jpg
+
+---
+
+# 读光文字识别
+## News
+- 2023年6月：
+    - 新增轻量化端侧识别[LightweightEdge-通用场景](https://www.modelscope.cn/models/damo/cv_LightweightEdge_ocr-recognitoin-general_damo/summary)模型和轻量化端侧[行检测模型](https://www.modelscope.cn/models/damo/cv_proxylessnas_ocr-detection-db-line-level_damo/summary)。
+- 2023年4月：
+    - 新增训练/微调时读取本地数据集的lmdb，用训练/微调后的模型继续识别，详见代码示例。
+- 2023年3月：
+    - 新增训练/微调流程，支持自定义参数及数据集，详见代码示例。
+- 2023年2月：
+    - 新增业界主流[CRNN-通用场景](https://www.modelscope.cn/models/damo/cv_crnn_ocr-recognition-general_damo/summary)模型。
+
+## 传送门
+各场景文本识别模型：
+- [ConvNextViT-手写场景](https://www.modelscope.cn/models/damo/cv_convnextTiny_ocr-recognition-handwritten_damo/summary)
+- [ConvNextViT-文档印刷场景](https://www.modelscope.cn/models/damo/cv_convnextTiny_ocr-recognition-document_damo/summary)
+- [ConvNextViT-自然场景](https://www.modelscope.cn/models/damo/cv_convnextTiny_ocr-recognition-scene_damo/summary)
+- [ConvNextViT-车牌场景](https://www.modelscope.cn/models/damo/cv_convnextTiny_ocr-recognition-licenseplate_damo/summary)
+- [CRNN-通用场景](https://www.modelscope.cn/models/damo/cv_crnn_ocr-recognition-general_damo/summary)
+- [LightweightEdge-通用场景](https://www.modelscope.cn/models/damo/cv_LightweightEdge_ocr-recognitoin-general_damo/summary)
+
+各场景文本检测模型：
+- [SegLink++-通用场景行检测](https://modelscope.cn/models/damo/cv_resnet18_ocr-detection-line-level_damo/summary)
+- [SegLink++-通用场景单词检测](https://modelscope.cn/models/damo/cv_resnet18_ocr-detection-word-level_damo/summary)
+- [DBNet-通用场景行检测](https://www.modelscope.cn/models/damo/cv_resnet18_ocr-detection-db-line-level_damo/summary)
+
+整图OCR能力：
+- [整图OCR-多场景](https://modelscope.cn/studios/damo/cv_ocr-text-spotting/summary)
+
+欢迎使用！
+
+## 模型描述
+- 文字识别，即给定一张文本图片，识别出图中所含文字并输出对应字符串。
+- 本模型主要包括三个主要部分，Convolutional Backbone提取图像视觉特征，ConvTransformer Blocks用于对视觉特征进行上下文建模，最后连接CTC loss进行识别解码以及网络梯度优化。识别模型结构如下图：   
+
+<p align="center">
+    <img src="./resources/ConvTransformer-Pipeline.jpg"/> 
+</p>
+
+## 期望模型使用方式以及适用范围
+本模型主要用于给输入图片输出图中文字内容，具体地，模型输出内容以字符串形式输出。用户可以自行尝试各种输入图片。具体调用方式请参考代码示例。
+- 注：输入图片应为包含文字的单行文本图片。其它如多行文本图片、非文本图片等可能没有返回结果，此时表示模型的识别结果为空。
+
+## 模型推理
+在安装完成ModelScope之后即可使用ocr-recognition的能力。(在notebook的CPU环境或GPU环境均可使用)
+- 使用图像的url，或准备图像文件上传至notebook（可拖拽）。
+- 输入下列代码。
+
+### 代码范例
+```python
+from modelscope.pipelines import pipeline
+from modelscope.utils.constant import Tasks
+import cv2
+
+ocr_recognition = pipeline(Tasks.ocr_recognition, model='damo/cv_convnextTiny_ocr-recognition-general_damo')
+
+### 使用url
+img_url = 'http://duguang-labelling.oss-cn-shanghai.aliyuncs.com/mass_img_tmp_20220922/ocr_recognition.jpg'
+result = ocr_recognition(img_url)
+print(result)
+
+### 使用图像文件
+### 请准备好名为'ocr_recognition.jpg'的图像文件
+# img_path = 'ocr_recognition.jpg'
+# img = cv2.imread(img_path)
+# result = ocr_recognition(img)
+# print(result)
+```
+
+### 模型可视化效果
+以下为模型的可视化文字识别效果。
+
+<p align="center">
+    <img src="./resources/rec_result_visu.jpg" width="400" /> 
+</p>
+
+### 模型局限性以及可能的偏差
+- 模型是在中英文数据集上训练的，在其他语言的数据上有可能产生一定偏差，请用户自行评测后决定如何使用。
+- 当前版本在python3.7的CPU环境和单GPU环境测试通过，其他环境下可用性待测试。
+
+## 模型微调/训练
+### 训练数据及流程介绍
+- 本文字识别模型训练数据集是MTWI以及部分收集数据，训练数据数量约6M。
+- 本模型参数随机初始化，然后在训练数据集上进行训练，在32x300尺度下训练20个epoch。
+
+### 模型微调/训练示例
+#### 训练数据集准备
+示例采用[ICDAR13手写数据集](https://modelscope.cn/datasets/damo/ICDAR13_HCTR_Dataset/summary)，已制作成lmdb，数据格式如下
+```
+'num-samples': number,
+'image-000000001': imagedata,
+'label-000000001': string,
+...
+```
+详情可下载解析了解。
+
+#### 配置训练参数并进行微调/训练
+参考代码及详细说明如下
+```python
+import os
+import tempfile
+
+from modelscope.hub.snapshot_download import snapshot_download
+from modelscope.metainfo import Trainers
+from modelscope.msdatasets import MsDataset
+from modelscope.trainers import build_trainer
+from modelscope.utils.config import Config, ConfigDict
+from modelscope.utils.constant import ModelFile, DownloadMode
+
+### 请确认您当前的modelscope版本，训练/微调流程在modelscope==1.4.0及以上版本中 
+### 当前notebook中版本为1.3.2，请手动更新，建议使用GPU环境
+
+model_id = 'damo/cv_convnextTiny_ocr-recognition-general_damo' 
+cache_path = snapshot_download(model_id) # 模型下载保存目录
+config_path = os.path.join(cache_path, ModelFile.CONFIGURATION) # 模型参数配置文件，支持自定义
+cfg = Config.from_file(config_path)
+
+# 构建数据集，支持自定义
+train_data_cfg = ConfigDict(
+    name='ICDAR13_HCTR_Dataset', 
+    split='test',
+    namespace='damo',
+    test_mode=False)
+
+train_dataset = MsDataset.load( 
+    dataset_name=train_data_cfg.name,
+    split=train_data_cfg.split,
+    namespace=train_data_cfg.namespace,
+    download_mode=DownloadMode.REUSE_DATASET_IF_EXISTS)
+
+test_data_cfg = ConfigDict(
+    name='ICDAR13_HCTR_Dataset',
+    split='test',
+    namespace='damo',
+    test_mode=True)
+
+test_dataset = MsDataset.load(
+    dataset_name=test_data_cfg.name,
+    split=test_data_cfg.split,
+    namespace=train_data_cfg.namespace,
+    download_mode=DownloadMode.REUSE_DATASET_IF_EXISTS)
+
+tmp_dir = tempfile.TemporaryDirectory().name # 模型文件和log保存位置，默认为"work_dir/"
+
+# 自定义参数，例如这里将max_epochs设置为15，所有参数请参考configuration.json
+def _cfg_modify_fn(cfg):
+    cfg.train.max_epochs = 15
+    return cfg
+
+####################################################################################
+
+'''
+使用本地文件
+    lmdb: 
+        构建包含下列信息的lmdb文件 (key: value)
+        'num-samples': 总样本数,
+        'image-000000001': 图像的二进制编码,
+        'label-000000001': 标签序列的二进制编码,
+        ...
+        image和label后的index为9位并从1开始
+下面为示例 (local_lmdb为本地的lmdb文件)
+'''
+
+# train_dataset = MsDataset.load( 
+#     dataset_name=train_data_cfg.name,
+#     split=train_data_cfg.split,
+#     namespace=train_data_cfg.namespace,
+#     download_mode=DownloadMode.REUSE_DATASET_IF_EXISTS,
+#     local_lmdb='./local_lmdb')
+
+# test_dataset = MsDataset.load(
+#     dataset_name=test_data_cfg.name,
+#     split=test_data_cfg.split,
+#     namespace=train_data_cfg.namespace,
+#     download_mode=DownloadMode.REUSE_DATASET_IF_EXISTS,
+#     local_lmdb='./local_lmdb')
+
+####################################################################################
+
+kwargs = dict(
+    model=model_id,
+    train_dataset=train_dataset,
+    eval_dataset=test_dataset,
+    work_dir=tmp_dir,
+    cfg_modify_fn=_cfg_modify_fn)
+
+# 模型训练
+trainer = build_trainer(name=Trainers.ocr_recognition, default_args=kwargs)
+trainer.train()
+```
+
+#### 用训练/微调后的模型进行识别
+```python
+from modelscope.pipelines import pipeline
+from modelscope.utils.constant import Tasks
+import os
+
+ep_num = 3  # 选择模型checkpoint
+cmd = 'cp {} {}'.format('./work_dir/epoch_%d.pth' % ep_num, './work_dir/output/pytorch_model.pt')  # 'work_dir'为configuration中设置的路径，'output'为输出默认路径
+os.system(cmd)
+ocr_recognition = pipeline(Tasks.ocr_recognition, model='./work_dir/output' )
+result = ocr_recognition('http://duguang-labelling.oss-cn-shanghai.aliyuncs.com/mass_img_tmp_20220922/ocr_recognition_icdar13.jpg')
+print(result)
+```

checkpoints/modelscope_ocr/damo/cv_convnextTiny_ocr-recognition-general_damo/configuration.json

@@ -0,0 +1,68 @@
+{
+    "framework": "pytorch",
+    "task": "ocr-recognition",
+    "pipeline": {
+        "type": "convnextTiny-ocr-recognition"
+    },
+    "model": {
+        "type": "OCRRecognition",
+        "recognizer": "ConvNextViT",
+        "inference_kwargs": {
+            "img_height": 32,
+            "img_width": 804,
+            "do_chunking": true
+        }        
+    },
+    "preprocessor": {
+        "type": "ocr-recognition"        
+    },
+    "train": {
+        "max_epochs": 30,
+        "work_dir": "./work_dir",
+        "dataloader": {
+            "batch_size_per_gpu": 32,
+            "workers_per_gpu": 0
+        },
+        "optimizer": {
+            "type": "AdamW",
+            "weight_decay": 0.01,
+            "lr": 0.001,
+            "options": {
+                "grad_clip": {
+                    "max_norm": 20
+                }
+            }
+        },
+        "lr_scheduler": {
+            "type": "MultiStepLR",
+            "milestones": [10, 20],
+            "gamma": 0.1
+        },
+        "hooks": [{
+                "type": "CheckpointHook",
+                "interval": 1,
+                "save_dir": "./work_dir"
+            },
+            {
+                "type": "TextLoggerHook",
+                "interval": 50,
+                "out_dir": "./work_dir"
+            },
+            {
+                "type": "IterTimerHook"
+            },
+            {
+                "type": "EvaluationHook",
+                "interval": 1
+            }
+        ]
+    },
+    "evaluation": {
+        "dataloader": {
+            "batch_size_per_gpu": 32,
+            "workers_per_gpu": 0,
+            "shuffle": false
+        },
+        "metrics": "ocr-recognition-metric"
+    }
+}

checkpoints/modelscope_ocr/damo/cv_convnextTiny_ocr-recognition-general_damo/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b4a624de1f099fede53622cf63e40fc07e2f09f1ef4e39d62bbc44858efdef4
+size 76859029

checkpoints/modelscope_ocr/damo/cv_convnextTiny_ocr-recognition-general_damo/pytorch_model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31cc4d17982aa82cfc7acfb95553b414183eaafafc7ff199692d4ebfbed7243c
+size 76836856

checkpoints/modelscope_ocr/damo/cv_convnextTiny_ocr-recognition-general_damo/quickstart.md

@@ -0,0 +1,161 @@
+---
+
+
+---
+## 模型推理
+- 使用图像的url，或准备图像文件。
+- 输入下列代码。
+
+### 代码范例
+```python
+from modelscope.pipelines import pipeline
+from modelscope.utils.constant import Tasks
+import cv2
+
+ocr_recognition = pipeline(Tasks.ocr_recognition, model='damo/cv_convnextTiny_ocr-recognition-general_damo')
+
+### 使用url
+img_url = 'http://duguang-labelling.oss-cn-shanghai.aliyuncs.com/mass_img_tmp_20220922/ocr_recognition.jpg'
+result = ocr_recognition(img_url)
+print(result)
+
+### 使用图像文件
+### 请准备好名为'ocr_recognition.jpg'的图像文件
+# img_path = 'ocr_recognition.jpg'
+# img = cv2.imread(img_path)
+# result = ocr_recognition(img)
+# print(result)
+```
+更多关于模型加载和推理的问题参考[模型的推理Pipeline](https://modelscope.cn/docs/%E6%A8%A1%E5%9E%8B%E7%9A%84%E6%8E%A8%E7%90%86Pipeline)。
+
+## 模型训练
+### 模型微调/训练示例
+#### 训练数据集准备
+示例采用[ICDAR13手写数据集](https://modelscope.cn/datasets/damo/ICDAR13_HCTR_Dataset/summary)，已制作成lmdb，数据格式如下
+```
+'num-samples': number,
+'image-000000001': imagedata,
+'label-000000001': string,
+...
+```
+详情可下载解析了解。
+
+#### 配置训练参数并进行微调/训练
+参考代码及详细说明如下
+```python
+import os
+import tempfile
+
+from modelscope.hub.snapshot_download import snapshot_download
+from modelscope.metainfo import Trainers
+from modelscope.msdatasets import MsDataset
+from modelscope.trainers import build_trainer
+from modelscope.utils.config import Config, ConfigDict
+from modelscope.utils.constant import ModelFile, DownloadMode
+
+### 请确认您当前的modelscope版本，训练/微调流程在modelscope==1.4.0及以上版本中 
+
+model_id = 'damo/cv_convnextTiny_ocr-recognition-general_damo' 
+cache_path = snapshot_download(model_id) # 模型下载保存目录
+config_path = os.path.join(cache_path, ModelFile.CONFIGURATION) # 模型参数配置文件，支持自定义
+cfg = Config.from_file(config_path)
+
+# 构建数据集，支持自定义
+train_data_cfg = ConfigDict(
+    name='ICDAR13_HCTR_Dataset', 
+    split='test',
+    namespace='damo',
+    test_mode=False)
+
+train_dataset = MsDataset.load( 
+    dataset_name=train_data_cfg.name,
+    split=train_data_cfg.split,
+    namespace=train_data_cfg.namespace,
+    download_mode=DownloadMode.REUSE_DATASET_IF_EXISTS)
+
+test_data_cfg = ConfigDict(
+    name='ICDAR13_HCTR_Dataset',
+    split='test',
+    namespace='damo',
+    test_mode=True)
+
+test_dataset = MsDataset.load(
+    dataset_name=test_data_cfg.name,
+    split=test_data_cfg.split,
+    namespace=train_data_cfg.namespace,
+    download_mode=DownloadMode.REUSE_DATASET_IF_EXISTS)
+
+tmp_dir = tempfile.TemporaryDirectory().name # 模型文件和log保存位置，默认为"work_dir/"
+
+# 自定义参数，例如这里将max_epochs设置为15，所有参数请参考configuration.json
+def _cfg_modify_fn(cfg):
+    cfg.train.max_epochs = 15
+    return cfg
+
+####################################################################################
+
+'''
+使用本地文件
+    lmdb: 
+        构建包含下列信息的lmdb文件 (key: value)
+        'num-samples': 总样本数,
+        'image-000000001': 图像的二进制编码,
+        'label-000000001': 标签序列的二进制编码,
+        ...
+        image和label后的index为9位并从1开始
+下面为示例 (local_lmdb为本地的lmdb文件)
+'''
+
+# train_dataset = MsDataset.load( 
+#     dataset_name=train_data_cfg.name,
+#     split=train_data_cfg.split,
+#     namespace=train_data_cfg.namespace,
+#     download_mode=DownloadMode.REUSE_DATASET_IF_EXISTS,
+#     local_lmdb='./local_lmdb')
+
+# test_dataset = MsDataset.load(
+#     dataset_name=test_data_cfg.name,
+#     split=test_data_cfg.split,
+#     namespace=train_data_cfg.namespace,
+#     download_mode=DownloadMode.REUSE_DATASET_IF_EXISTS,
+#     local_lmdb='./local_lmdb')
+
+####################################################################################
+
+kwargs = dict(
+    model=model_id,
+    train_dataset=train_dataset,
+    eval_dataset=test_dataset,
+    work_dir=tmp_dir,
+    cfg_modify_fn=_cfg_modify_fn)
+
+# 模型训练
+trainer = build_trainer(name=Trainers.ocr_recognition, default_args=kwargs)
+trainer.train()
+```
+
+#### 用训练/微调后的模型进行识别
+```python
+from modelscope.pipelines import pipeline
+from modelscope.utils.constant import Tasks
+import os
+
+ep_num = 3  # 选择模型checkpoint
+cmd = 'cp {} {}'.format('./work_dir/epoch_%d.pth' % ep_num, './work_dir/output/pytorch_model.pt')  # 'work_dir'为configuration中设置的路径，'output'为输出默认路径
+os.system(cmd)
+ocr_recognition = pipeline(Tasks.ocr_recognition, model='./work_dir/output' )
+result = ocr_recognition('http://duguang-labelling.oss-cn-shanghai.aliyuncs.com/mass_img_tmp_20220922/ocr_recognition_icdar13.jpg')
+print(result)
+```
+更多使用说明请参阅[ModelScope文档中心](http://www.modelscope.cn/#/docs)。
+
+---
+
+
+---
+## 下载并安装ModelScope library
+更多关于下载安装ModelScope library的问题参考[环境安装](https://modelscope.cn/docs/%E7%8E%AF%E5%A2%83%E5%AE%89%E8%A3%85)。
+
+```python
+pip install "modelscope[cv]" -f https://modelscope.oss-cn-beijing.aliyuncs.com/releases/repo.html
+```

checkpoints/modelscope_ocr/damo/cv_convnextTiny_ocr-recognition-general_damo/vocab.txt

@@ -0,0 +1,7643 @@
+在
+保
+证
+商
+品
+和
+外
+包
+装
+完
+好
+无
+损
+情
+况
+下
+与
+本
+店
+客
+服
+联
+系
+并
+百
+代
+采
+购
+周
+期
+为
+1
+-
+3
+天
+二
+检
+测
+了
+提
+香
+。
+单
+项
+深
+度
+法
+的
+目
+标
+艺
+术
+对
+其
+他
+发
+展
+领
+域
+促
+进
+五
+虚
+伪
+人
+味
+及
+雅
+量
+第
+十
+章
+袁
+世
+凯
+你
+何
+成
+不
+华
+盛
+顿
+一
+精
+塑
+封
+因
+此
+可
+以
+得
+出
+两
+个
+结
+论
+是
+大
+道
+养
+生
+中
+医
+西
+都
+拒
+绝
+没
+有
+能
+具
+备
+力
+朵
+莲
+花
+按
+照
+湿
+地
+态
+统
+更
+多
+新
+款
+欢
+迎
+请
+加
+主
+林
+动
+眼
+神
+阴
+沉
+望
+着
+那
+夹
+杂
+狂
+暴
+波
+掠
+而
+来
+元
+光
+芒
+电
+子
+产
+较
+手
+笔
+京
+城
+饭
+顶
+层
+旋
+转
+餐
+厅
+绞
+股
+蓝
+抗
+衰
+老
+、
+预
+防
+肿
+瘤
+抑
+制
+溃
+疡
+缓
+解
+紧
+张
+疲
+劳
+镇
+蛋
+糕
+配
+套
+赠
+椰
+滚
+落
+水
+时
+又
+恰
+巧
+砸
+晕
+正
+旁
+边
+看
+热
+闹
+鱼
+陈
+荣
+茂
+帮
+年
+轻
+朋
+友
+朝
+气
+蓬
+勃
+善
+于
+学
+习
+勇
+创
+顾
+阑
+珊
+刚
+走
+门
+7
+.
+2
+设
+计
+药
+形
+局
+限
+性
+社
+区
+心
+理
+康
+复
+家
+庭
+七
+我
+们
+否
+应
+该
+停
+忙
+碌
+脚
+步
+唉
+它
+说
+过
+朴
+素
+【
+介
+绍
+】
+刑
+公
+感
+(
+江
+苏
+)
+买
+日
+0
+6
+8
+挺
+P
+r
+e
+n
+a
+t
+l
+N
+u
+i
+o
+d
+f
+s
+几
+乎
+被
+噩
+运
+溺
+毙
+她
+后
+化
+作
+李
+源
+儿
+再
+回
+到
+非
+之
+付
+红
+屁
+疹
+4
+内
+容
+识
+别
+带
+风
+扇
+机
+纯
+铜
+高
+梁
+强
+措
+施
+合
+安
+排
+受
+方
+式
+书
+布
+朗
+小
+《
+使
+魔
+鬼
+》
+A
+g
+研
+究
+三
+节
+独
+山
+石
+分
+司
+全
+面
+算
+管
+用
+实
+例
+四
+国
+际
+贸
+易
+划
+程
+骤
+类
+型
+这
+散
+文
+概
+念
+#
+S
+w
+p
+C
+m
+钢
+膜
+屏
+覆
+盖
+炉
+铁
+口
+液
+压
+开
+钻
+杆
+自
+注
+浆
+锚
+体
+9
+5
+,
+h
+c
+v
+B
+初
+次
+或
+长
+放
+置
+定
+要
+充
+会
+查
+验
+库
+存
+通
+知
+您
+货
+;
+月
+O
+海
+拔
+:
+均
+超
+临
+床
+必
+读
+丛
+厘
+摩
+Y
+G
+R
+毫
+克
+镁
+码
+关
+九
+官
+选
+士
+举
+土
+工
+立
+纹
+饰
+从
+群
+治
+疗
+农
+业
+技
+校
+讲
+述
+上
+质
+错
+邮
+浙
+川
+贵
+州
+云
+南
+藏
+也
+叫
+邦
+义
+军
+政
+重
+确
+牛
+皮
+癣
+症
+经
+组
+减
+观
+剧
+须
+戏
+捧
+教
+版
+课
+准
+依
+据
+俄
+普
+希
+金
+|
+编
+燕
+译
+者
+吕
+捷
+绘
+画
+宏
+亚
+图
+双
+显
+卡
+智
+切
+换
+偿
+乙
+肝
+炎
+硬
+患
+还
+啦
+简
+已
+车
+胶
+清
+洁
+位
+直
+接
+黏
+即
+号
+适
+哦
+纸
+I
+D
+瑞
+丽
+志
+德
+伯
+特
+灵
+格
+蔡
+部
+温
+控
+器
+优
+势
+吴
+景
+横
+晖
+郭
+玉
+费
+卖
+各
+承
+担
+半
+假
+若
+键
+刻
+醒
+每
+种
+只
+室
+入
+处
+事
+美
+恩
+房
+龙
+著
+晏
+榕
+莫
+尔
+杜
+霍
+维
+奇
+英
+怀
+杨
+富
+斌
+最
+修
+订
+权
+威
+登
+舞
+银
+牌
+级
+*
+y
+z
+V
+/
+H
+J
+k
+专
+样
+前
+试
+剂
+盒
+府
+奖
+持
+宝
+肤
+常
+健
+基
+础
+指
+短
+线
+操
+旅
+游
+纪
+币
+间
+价
+值
+涵
+环
+境
+审
+物
+畅
+销
+影
+响
+启
+蒙
+虽
+然
+答
+拉
+里
+料
+故
+誓
+浪
+漫
+燃
+烧
+脏
+平
+随
+呈
+增
+太
+阳
+察
+护
+T
+M
+尺
+表
+某
+件
+衣
+寸
+参
+考
+六
+角
+螺
+栓
+固
+盗
+给
+籍
+借
+钱
+破
+坏
+融
+秩
+序
+罪
+律
+责
+任
+民
+院
+反
+倾
+行
+案
+干
+问
+题
+规
++
+植
+绒
+背
+板
+b
+需
+向
+咨
+陶
+瓷
+球
+油
+润
+滑
+属
+擦
+数
+L
+韩
+谨
+骗
+推
+荐
+喜
+洋
+底
+状
+让
+延
+握
+柄
+滤
+雾
+℃
+氧
+火
+原
+杏
+仁
+芝
+迷
+E
+埃
+汇
+韵
+达
+等
+快
+递
+如
+添
+珍
+惜
+藻
+透
+明
+W
+?
+=
+拍
+岩
+鲜
+汀
+永
+杭
+微
+~
+K
+速
+浏
+览
+F
+千
+米
+示
+宽
+比
+余
+条
+务
+语
+秒
+连
+续
+信
+息
+U
+堂
+般
+蚝
+鸡
+浅
+折
+叠
+虞
+劲
+松
+拐
+擎
+杰
+斐
+逊
+x
+黄
+暗
+蒋
+猜
+谜
+宣
+言
+交
+锋
+词
+省
+略
+果
+泥
+渣
+砖
+砌
+块
+造
+妇
+罗
+满
+佳
+八
+鸿
+運
+木
+材
+演
+經
+典
+風
+刮
+壁
+星
+搅
+拌
+乳
+卷
+晋
+胡
+萝
+卜
+饼
+哭
+吧
+j
+将
+芯
+哪
+些
+贝
+惠
+券
+命
+名
+则
+密
+佛
+珠
+串
+%
+隔
+兰
+引
+孩
+铝
+架
+男
+女
+古
+鲁
+迅
+逆
+耳
+话
+忠
+效
+办
+肠
+便
+秘
+孕
+妈
+幽
+默
+打
+扉
+阅
+册
+循
+渐
+字
+喻
+贫
+贡
+献
+界
+移
+现
+挑
+战
+流
+网
+络
+误
+功
+点
+认
+色
+彩
+磁
+路
+构
+矢
+场
+泄
+泻
+园
+建
+变
+频
+冻
+桥
+震
+害
+企
+含
+符
+Q
+片
+污
+染
+析
+模
+拟
+X
+集
+供
+链
+粒
+袋
+相
+耕
+械
+趋
+活
+资
+市
+吞
+吐
+投
+险
+访
+同
+济
+仲
+裁
+诉
+讼
+核
+极
+晶
+输
+曲
+获
+取
+助
+监
+由
+团
+户
+账
+弧
+附
+帐
+汗
+调
+慢
+跳
+拦
+综
+征
+矫
+函
+箱
+血
+糖
+仪
+做
+细
+径
+档
+先
+声
+音
+冷
+凝
+称
+当
+才
+支
+赔
+寄
+磕
+碰
+难
+免
+掉
+漆
+敬
+��
+毒
+夜
+雨
+避
+缝
+吸
+盘
+议
+坐
+姿
+所
+售
+翡
+翠
+镯
+缅
+甸
+改
+织
+割
+截
+圆
+峰
+就
+沪
+首
+斤
+起
+墨
+粘
+斯
+矛
+盾
+意
+盟
+玻
+璃
+汁
+瓶
+洗
+忧
+笑
+赛
+乃
+姆
+灌
+缩
+衷
+季
+酸
+酵
+氨
+纶
+聚
+酯
+纤
+差
+空
+伸
+弯
+辑
+估
+乐
+票
+椭
+令
+拓
+训
+练
+倪
+思
+博
+传
+策
+把
+己
+茶
+段
+冲
+醚
+妙
+填
+趣
+@
+妆
+沐
+浴
+露
+吊
+Z
+燈
+圓
+伞
+狐
+狸
+毛
+断
+降
+恢
+亮
+颗
+离
+孔
+钟
+跑
+消
+耗
+缘
+阻
+吉
+虎
+鼠
+共
+鹰
+锈
+柱
+蜗
+亲
+索
+求
+评
+凭
+广
+告
+墅
+焊
+悬
+臂
+整
+堆
+未
+梳
+摄
+菌
+菇
+棚
+夏
+印
+卫
+批
+候
+送
+拼
+硼
+税
+视
+暖
+轴
+镜
+执
+粗
+―
+丈
+夫
+职
+田
+纳
+慕
+宁
+静
+万
+枚
+饿
+蛇
+桢
+鸵
+鸟
+栗
+桂
+炖
+猪
+蹄
+[
+]
+扰
+棒
+稳
+零
+师
+葱
+什
+么
+退
+协
+利
+真
+遗
+头
+记
+脑
+升
+互
+溫
+莎
+爵
+愛
+象
+徵
+很
+瀑
+怎
+骨
+击
+奏
+弘
+師
+開
+剑
+休
+颜
+耐
+蕴
+铀
+列
+枕
+醋
+敷
+雷
+萨
+轿
+疑
+障
+汽
+缸
+棕
+浦
+伦
+敦
+食
+伤
+身
+圣
+迹
+稀
+青
+巅
+雄
+潤
+灯
+'
+白
+豹
+桶
+另
+营
+率
+爆
+绳
+扣
+丸
+睡
+遣
+削
+弱
+异
+壮
+收
+颈
+椎
+病
+吗
+酱
+根
+宗
+盐
+焗
+筒
+晓
+珑
+辽
+兴
+扩
+幸
+福
+蒲
+听
+馆
+祭
+北
+谓
+臀
+肌
+挛
+归
+猴
+脱
+磷
+诺
+东
+县
+今
+邹
+村
+历
+涯
+副
+席
+央
+擅
+兽
+毕
+武
+汉
+科
+祝
+范
+泰
+斗
+哲
+兼
+闻
+许
+留
+至
+乌
+党
+派
+坪
+寒
+厂
+铃
+锐
+港
+航
+触
+座
+众
+钮
+尽
+致
+慎
+申
+丹
+欧
+筑
+赋
+媒
+低
+炮
+草
+橡
+顺
+丰
+总
+委
+决
+仿
+辅
+射
+钳
+锯
+锉
+去
+幕
+菜
+择
+软
+刀
+牙
+王
+终
+炒
+锅
+磨
+树
+脂
+浸
+阿
+巴
+舍
+抹
+灰
+录
+矩
+坝
+咖
+仓
+额
+紫
+磅
+积
+近
+似
+龄
+丝
+池
+颠
+谐
+谱
+阶
+报
+约
+偶
+写
+赏
+轨
+掺
+搞
+残
+导
+溶
+铅
+搭
+员
+契
+站
+围
+苯
+胺
+址
+尼
+谈
+判
+纵
+描
+像
+塞
+楼
+荷
+载
+锁
+烈
+诱
+欲
+垫
+骑
+寿
+失
+婚
+姻
+继
+诚
+腊
+庙
+予
+摇
+篮
+浮
+廉
+良
+宿
+句
+逻
+沟
+困
+涉
+询
+爱
+革
+找
+突
+嘉
+庆
+痛
+圳
+届
+洲
+足
+锦
+旗
+住
+惑
+秋
+冬
+枪
+萌
+闪
+熊
+猫
+衫
+杯
+伟
+纱
+童
+始
+刷
+雪
+纺
+厚
+针
+潮
+撞
+尚
+春
+棉
+麻
+恤
+侣
+薄
+拖
+鞋
+凉
+津
+肩
+闲
+羊
+巾
+黑
+胖
+裤
+腰
+肥
+袖
+衬
+裙
+豆
+胸
+翻
+唯
+舰
+框
+偏
+脸
+跟
+台
+嘴
+镂
+巡
+唱
+蝴
+蝶
+仔
+帽
+斜
+挎
+瘦
+熟
+轮
+煎
+艾
+蒿
+粑
+岁
+扭
+荧
+早
+享
+育
+苗
+马
+丁
+括
+遵
+守
+除
+疆
+讨
+兑
+涤
+钓
+椅
+渔
+竿
+兄
+弟
+煤
+跨
+屋
+檩
+恋
+仅
+莱
+勒
+泵
+寬
+長
+燥
+洞
+穴
+益
+靠
+父
+铺
+储
+倒
+株
+沿
+贴
+招
+剩
+剪
+凡
+霜
+梨
+涛
+页
+尤
+羽
+枣
+爬
+尾
+钠
+钾
+蒸
+镀
+宰
+辨
+娃
+哈
+窟
+少
+追
+疸
+焦
+胆
+母
+玩
+壳
+纲
+末
+驱
+衡
+殊
+垮
+栈
+嵌
+粉
+碎
+租
+赁
+贷
+烟
+囱
+族
+绩
+涡
+垂
+隐
+俗
+左
+右
+篇
+陌
+讪
+拿
+颖
+绿
+酒
+犯
+既
+遂
+队
+胃
+厕
+敌
+忆
+坤
+培
+蜚
+絮
+飘
+萍
+懒
+柴
+董
+昌
+昼
+摊
+居
+携
+郑
+爽
+狗
+漏
+束
+脊
+补
+齐
+端
+窥
+往
+①
+叶
+琴
+捕
+鲫
+腐
+蚀
+凹
+坑
+裂
+缺
+陷
+混
+挽
+觉
+阔
+想
+钩
+「
+」
+癫
+痫
+释
+妖
+怪
+氏
+!
+插
+杀
+卉
+争
+恶
+剖
+阀
+帜
+探
+圈
+摸
+坡
+份
+崎
+飞
+锤
+歌
+拧
+懂
+沈
+忽
+悠
+钉
+闭
+晚
+播
+菩
+愿
+锂
+佬
+吹
+貌
+骄
+傲
+麦
+蒂
+幻
+裹
+侦
+脖
+驾
+楠
+柜
+_
+脉
+寻
+托
+宜
+撕
+恨
+脆
+绚
+禁
+忌
+绣
+穿
+痕
+浓
+熔
+详
+碳
+酶
+滨
+仙
+途
+铣
+蹈
+淘
+隧
+饱
+疏
+勾
+阵
+瞻
+远
+瞩
+诀
+攻
+踪
+灾
+森
+允
+越
+炼
+距
+竞
+喀
+旭
+寨
+玫
+瑰
+烤
+礼
+肉
+叹
+拆
+砝
+野
+炸
+驴
+慰
+船
+勘
+井
+够
+勿
+掌
+泽
+碑
+仕
+勺
+奶
+虹
+抱
+午
+罐
+渴
+抵
+弃
+负
+遇
+贯
+窗
+债
+哥
+虫
+氢
+榜
+赢
+矿
+河
+噪
+漂
+拥
+宇
+宙
+髓
+柳
+菟
+桃
+芽
+旬
+践
+弹
+趾
+梦
+镶
+抓
+眨
+睛
+隙
+久
+咱
+淋
+锌
+帧
+糊
+挤
+诊
+辈
+淀
+催
+帘
+蚊
+挂
+妻
+矮
+姓
+奕
+卧
+刃
+侧
+抉
+硅
+努
+舒
+铍
+急
+疾
+蜂
+蜇
+齿
+酰
+碱
+瓣
+救
+冗
+槽
+讯
+侵
+占
+财
+绕
+啤
+虑
+烘
+焙
+魅
+★
+竹
+捏
+惯
+靶
+尿
+纫
+沥
+遥
+穷
+甲
+烯
+卒
+危
+敏
+褪
+坠
+铂
+腭
+戒
+缠
+绵
+恒
+耀
+鲤
+警
+泡
+節
+孫
+寶
+琦
+續
+與
+談
+吃
+抬
+栅
+溢
+塔
+戴
+坚
+抽
+袜
+皿
+胁
+且
+钞
+帅
+碧
+柔
+砂
+膏
+q
+豪
+蝙
+蝠
+佩
+醫
+醉
+淡
+诗
+櫻
+曼
+頓
+诞
+栽
+眉
+檀
+桑
+柑
+橘
+晨
+Ⅳ
+授
+绽
+姐
+胰
+腺
+钛
+摆
+Ⅲ
+刺
+乡
+粮
+榨
+筋
+箍
+墙
+禧
+驰
+骋
+烷
+\
+愉
+蔬
+蔻
+犬
+宠
+悦
+吟
+△
+映
+灸
+耍
+$
+洽
+玛
+瑙
+待
+厌
+烦
+橙
+旦
+抄
+袭
+雕
+班
+钙
+肽
+魚
+為
+稱
+對
+體
+類
+呢
+氩
+弦
+私
+街
+叉
+��
+赵
+谷
+弊
+辐
+肚
+兜
+毯
+屉
+洛
+徒
+奥
+筹
+祈
+振
+朱
+宋
+舆
+胜
+穆
+洙
+严
+佟
+签
+廊
+迈
+戈
+皇
+帝
+菱
+婴
+骆
+驼
+帆
+赶
+碗
+靴
+葩
+嫦
+娥
+虾
+汤
+腾
+渊
+苹
+尖
+時
+盏
+幼
+珂
+吻
+芙
+妮
+陆
+囊
+摔
+膀
+胱
+唇
+硕
+揽
+旧
+{
+}
+旺
+※
+Ⅵ
+②
+趺
+禅
+腿
+③
+④
+弗
+紀
+庫
+顧
+問
+創
+辦
+許
+■
+▲
+▼
+谢
+累
+◆
+坊
+◇
+暂
+○
+◎
+邪
+惬
+●
+撼
+萧
+咪
+裱
+☆
+冠
+亏
+唐
+娜
+史
+楚
+眸
+葛
+贤
+劝
+箴
+伺
+啊
+翅
+怨
+叙
+止
+瘢
+挥
+豇
+喷
+泉
+绸
+涩
+乱
+□
+牧
+谁
+咕
+噜
+昏
+焓
+隼
+阐
+姑
+娘
+丫
+鬟
+肢
+腹
+饮
+彼
+潘
+烹
+饪
+烃
+準
+斷
+進
+場
+點
+訣
+竅
+逢
+奴
+赞
+呵
+悉
+浩
+劫
+辟
+冰
+淇
+刊
+抢
+娱
+踏
+夸
+慧
+婉
+腔
+稍
+聘
+鲑
+卵
+峻
+谣
+怠
+歉
+町
+诫
+痰
+郁
+脾
+昧
+棱
+崇
+祯
+却
+晰
+腻
+扬
+麟
+昆
+韭
+椒
+雍
+疯
+粳
+稻
+獵
+繼
+尋
+莉
+悟
+辛
+苦
+兔
+仗
+碼
+丑
+柯
+芳
+苓
+绎
+哀
+枭
+缔
+迄
+颇
+朽
+佶
+撰
+蕾
+療
+術
+務
+陕
+踢
+净
+恐
+慌
+秀
+滔
+讳
+绪
+帕
+刘
+吾
+谚
+眠
+鼾
+死
+姜
+辣
+疼
+後
+愚
+疫
+脐
+翁
+環
+機
+懷
+裡
+芬
+佣
+沙
+漠
+沸
+痹
+澳
+丙
+碟
+靓
+殖
+阈
+褐
+虱
+泪
+搜
+窑
+岸
+耦
+祖
+乘
+沃
+撑
+缴
+筛
+魄
+膨
+胀
+繁
+茎
+贾
+屠
+岳
+拳
+適
+獨
+裝
+設
+計
+會
+來
+聆
+亡
+鸦
+埠
+啡
+诈
+祥
+煞
+倍
+堵
+橱
+伴
+禽
+緻
+欣
+诵
+曾
+皂
+祛
+斑
+跌
+毅
+剃
+僧
+返
+糟
+屈
+寂
+寞
+劣
+癌
+滿
+氛
+圍
+秉
+栋
+襟
+辩
+废
+邀
+泼
+梯
+镍
+踩
+栏
+绊
+迟
+稿
+窍
+卢
+焕
+君
+姚
+嘲
+讽
+畸
+巨
+棋
+梅
+尹
+鹅
+杉
+薛
+玄
+爸
+帖
+亿
+忍
+佚
+谦
+粤
+涂
+纽
+茨
+韦
+剔
+廷
+瑶
+谛
+枸
+杞
+薯
+龟
+羲
+亭
+『
+』
+堇
+現
+貨
+厨
+缚
+娇
+猿
+瓦
+楞
+宫
+彬
+玳
+瑁
+約
+義
+佰
+喇
+叭
+荡
+拘
+窄
+彦
+俱
+忘
+迪
+锆
+馨
+煲
+睿
+岛
+藕
+铬
+遮
+亥
+肖
+湖
+〔
+〕
+扛
+冒
+〓
+甘
+録
+岗
+卤
+坦
+唾
+召
+贽
+纠
+纷
+饲
+筮
+說
+於
+陽
+謂
+凤
+贺
+嫁
+粥
+伙
+肾
+俊
+傻
+徽
+庐
+徐
+泛
+尘
+孽
+慈
+悲
+但
+俯
+恭
+栩
+鹿
+妥
+肯
+悔
+辞
+愁
+扯
+彻
+甩
+泊
+饥
+狼
+箔
+剥
+钴
+败
+垛
+楂
+枝
+甚
+辆
+嗅
+疤
+鉴
+幅
+誉
+孝
+贩
+欠
+庄
+轭
+勤
+呼
+遍
+栖
+泌
+羌
+硝
+盆
+鼻
+葡
+萄
+酷
+炫
+浒
+峡
+寇
+颉
+蓄
+猎
+卸
+極
+數
+總
+邻
+拾
+喂
+鸭
+￥
+昕
+凿
+兵
+妃
+瞧
+哟
+敲
+捆
+亦
+哉
+乏
+霉
+喝
+搓
+將
+罩
+揉
+尴
+尬
+敢
+艰
+萃
+篷
+拎
+稠
+俭
+翼
+岭
+笼
+垃
+圾
+庞
+雀
+嘱
+姬
+铭
+笛
+葫
+芦
+杠
+臭
+贪
+贿
+赂
+滞
+奠
+扳
+皆
+莞
+丢
+衔
+〗
+丧
+励
+逃
+侍
+湘
+霸
+诸
+魂
+傩
+驻
+昊
+湾
+噎
+澡
+弄
+轩
+祸
+迫
+睫
+斓
+壶
+秤
+伍
+搬
+怡
+替
+谋
+嗣
+乔
+胳
+膊
+羞
+炔
+庸
+匡
+乞
+瑕
+疵
+怕
+驭
+珀
+敛
+讓
+戀
+蠢
+陋
+巩
+押
+番
+茄
+狱
+曹
+兹
+寓
+惊
+畜
+葵
+铸
+伽
+瑜
+柬
+埔
+歙
+砚
+魏
+偷
+烛
+馈
+督
+咽
+胎
+酪
+黎
+跃
+蹬
+簧
+瘾
+沂
+贮
+砍
+颌
+鸥
+彈
+詞
+輯
+臣
+孙
+蓉
+赚
+釉
+溪
+泸
+沽
+剿
+捻
+沫
+匀
+巢
+秦
+竣
+矾
+爹
+锡
+溧
+辗
+鼓
+莆
+褥
+奉
+藉
+颅
+糯
+咯
+惧
+黛
+孵
+哺
+刹
+麼
+縮
+們
+沒
+餘
+仆
+劈
+骂
+傅
+傍
+疮
+赖
+耿
+籽
+仍
+殿
+卓
+俞
+滇
+缤
+尧
+沛
+奔
+裸
+尊
+喊
+奋
+赤
+聪
+瞎
+烂
+掩
+揭
+畏
+扎
+涅
+诲
+披
+稚
+盈
+遭
+琢
+孤
+猩
+盔
+菲
+翰
+鲍
+秆
+蔗
+甜
+颤
+抖
+蚱
+蜢
+撒
+蟾
+逾
+泳
+鳖
+拢
+梭
+灿
+娴
+淑
+佑
+荒
+曝
+捣
+捞
+拣
+呆
+堪
+聋
+砭
+挟
+逼
+枫
+挡
+挚
+酣
+漓
+茫
+喵
+菊
+桡
+疝
+璧
+昨
+藤
+蔓
+莓
+伊
+厄
+瓜
+绅
+谅
+祺
+剌
+潜
+毁
+鼎
+砰
+忐
+忑
+饕
+餮
+惆
+怅
+竟
+鞍
+壤
+晃
+闷
+屡
+拜
+舌
+噢
+宅
+缎
+焚
+苍
+煌
+钧
+栾
+彭
+麒
+泓
+筱
+汪
+陵
+鹏
+亨
+御
+瑛
+蕊
+辉
+啥
+吼
+激
+蹭
+岂
+蹋
+砒
+蜜
+汰
+晾
+尝
+嗯
+彰
+掰
+拭
+彝
+嫩
+宛
+枯
+萎
+晒
+淫
+梗
+俩
+桩
+鑫
+钰
+榴
+皓
+罚
+國
+儒
+統
+荟
+搏
+笨
+烙
+腕
+锃
+袍
+筝
+肺
+悼
+陪
+酬
+憧
+憬
+盲
+罢
+涨
+艳
+俪
+渠
+滴
+蝼
+蛄
+卦
+趟
+咬
+鳞
+禄
+箫
+蝎
+丘
+厦
+侄
+畴
+牢
+蔽
+挖
+掘
+倡
+摹
+凸
+崖
+柏
+弥
+缆
+肃
+撮
+翌
+讹
+妨
+冤
+羔
+郎
+狭
+颂
+凱
+亞
+瑟
+扫
+茅
+纬
+夺
+抒
+犹
+蜀
+灭
+弓
+狮
+胞
+札
+啸
+瑚
+违
+伏
+蛙
+鸣
+吠
+澤
+镖
+舟
+烨
+晟
+滋
+翘
+绢
+牡
+璞
+腌
+瞬
+搁
+岱
+旨
+氯
+辄
+刨
+朕
+鲨
+媛
+琳
+凳
+韧
+匠
+逸
+愧
+聊
+昭
+颁
+勋
+篆
+肪
+凌
+扮
+宦
+逝
+拽
+哼
+凰
+灶
+洪
+揣
+匹
+叮
+锻
+墓
+鹤
+渍
+蛀
+颐
+琉
+耻
+辱
+暇
+籁
+掀
+怜
+惨
+呀
+戛
+翩
+膳
+娩
+愈
+坯
+逐
+醇
+坛
+喽
+赫
+蛤
+惫
+鞑
+虏
+橄
+榄
+斋
+舜
+昂
+扶
+炜
+裴
+迁
+昙
+曜
+坎
+狠
+诿
+���
+叔
+柠
+檬
+喱
+酥
+咦
+咳
+咸
+玲
+堡
+驗
+際
+賽
+哆
+雇
+泣
+哮
+喘
+奸
+唤
+惰
+啃
+啄
+喔
+矶
+嗜
+嘿
+夕
+蘑
+雁
+熨
+烫
+專
+賣
+規
+衛
+題
+離
+苛
+炬
+捉
+拨
+糙
+雏
+轧
+愤
+怒
+驚
+匣
+湮
+仄
+窝
+埋
+撤
+匿
+棍
+泾
+垓
+垒
+宪
+髋
+墟
+殷
+拂
+壬
+癸
+馏
+蒜
+甫
+摁
+抚
+皙
+阙
+寡
+殆
+绑
+阱
+慵
+︰
+拙
+坟
+舀
+钗
+爷
+蚕
+盂
+吨
+採
+購
+優
+妞
+碣
+蜡
+阁
+佐
+镰
+乖
+谥
+吓
+桐
+豫
+凑
+扔
+坞
+雯
+悚
+妲
+侯
+诣
+婆
+娑
+滕
+裳
+孟
+羡
+蹴
+顽
+痒
+锥
+酚
+圭
+轼
+汴
+瞄
+绶
+宾
+耶
+猛
+邯
+扁
+脯
+豁
+堤
+樱
+歇
+援
+曳
+實
+繪
+編
+诙
+淆
+熙
+霖
+煮
+渗
+惕
+測
+間
+癒
+妩
+媚
+琥
+憨
+暄
+叛
+踵
+牲
+厉
+嘶
+膛
+毋
+瀚
+蔷
+薇
+撬
+屯
+崔
+嵇
+邢
+攒
+髦
+潭
+祁
+履
+阎
+迦
+牟
+挪
+萤
+嫉
+妒
+烁
+塘
+貔
+貅
+狄
+郊
+驹
+溥
+侗
+霆
+簸
+崩
+簿
+挣
+虔
+浑
+蛮
+鞭
+惹
+埂
+妹
+庵
+裔
+恕
+恳
+悍
+奢
+姨
+恼
+隆
+兆
+贰
+樵
+酝
+酿
+哒
+枉
+躯
+晴
+嘘
+憾
+勉
+窜
+殉
+浊
+踮
+蹑
+攀
+艇
+戳
+兩
+據
+資
+悸
+罄
+痧
+伶
+仃
+桌
+挫
+颦
+惮
+翱
+翔
+抛
+膝
+谭
+拷
+疚
+镐
+赃
+恪
+討
+搪
+郾
+跤
+擁
+钥
+匙
+谊
+隅
+敕
+诛
+诩
+學
+種
+選
+處
+挝
+迭
+羚
+戚
+跆
+囚
+奈
+爪
+扑
+潺
+斧
+毂
+瓢
+漳
+茬
+摘
+逵
+赴
+叵
+募
+俏
+寺
+曦
+洒
+阜
+霞
+叁
+睐
+趁
+卑
+坷
+醐
+弛
+窃
+辫
+牺
+盼
+決
+戰
+驶
+質
+渡
+垢
+镕
+槿
+钝
+蚂
+蚁
+熏
+钊
+氟
+幂
+澄
+隋
+炀
+⑤
+倦
+晦
+曙
+靖
+璇
+槃
+渎
+豚
+拯
+芭
+夾
+頭
+車
+細
+軸
+桔
+冶
+梵
+碍
+棵
+樊
+歸
+園
+書
+盎
+乒
+乓
+⊙
+視
+∕
+稽
+歪
+異
+終
+湛
+榭
+個
+誤
+屬
+魁
+痂
+毡
+鳝
+氦
+氣
+溴
+氮
+沾
+昵
+媲
+娟
+淮
+辙
+皖
+滩
+佤
+冕
+崛
+沦
+挞
+骏
+淳
+泗
+磬
+曰
+榆
+嵩
+炭
+澜
+蜃
+涓
+淝
+淞
+涮
+淺
+渤
+薰
+滥
+觞
+贻
+侨
+發
+積
+烽
+蕨
+棠
+庇
+牵
+萦
+犀
+猕
+馄
+饨
+悄
+嘻
+猶
+徘
+徊
+遠
+過
+傳
+奘
+怔
+诘
+綦
+跋
+汲
+尉
+浇
+漾
+腑
+嗽
+倚
+瑾
+俘
+Ｖ
+侠
+咩
+苔
+辰
+孚
+觀
+亂
+厢
+槛
+盅
+Ｔ
+宴
+闵
+荤
+脍
+炙
+當
+這
+樣
+贱
+瘟
+桦
+蛾
+灼
+璐
+褶
+伐
+﹐
+胥
+裕
+翳
+疔
+錄
+稣
+仇
+瞿
+畔
+鸠
+粪
+娌
+狩
+売
+菠
+屐
+藜
+厥
+贼
+芥
+耘
+龛
+钦
+爾
+動
+斩
+棘
+镳
+荨
+嫌
+舅
+欺
+妾
+谏
+痴
+凶
+恃
+役
+刁
+狡
+赐
+绫
+寝
+∶
+乾
+冥
+腦
+蹲
+璀
+璨
+冀
+罹
+殃
+靈
+詳
+粱
+瘫
+赎
+袈
+裟
+辖
+弈
+逞
+銮
+诏
+熬
+硒
+捐
+邂
+逅
+貂
+骷
+髅
+蝉
+谍
+寥
+娅
+給
+話
+詹
+砺
+趴
+汾
+妳
+绛
+矣
+仰
+秽
+侬
+狙
+翟
+捡
+喋
+仑
+咏
+麾
+琐
+赣
+蝇
+蟹
+胚
+婀
+绺
+腋
+拗
+粹
+胫
+哄
+芊
+皱
+樂
+笋
+檐
+殡
+葬
+苁
+戟
+苑
+苟
+戊
+岚
+溜
+萱
+莠
+莴
+苣
+荸
+荠
+陀
+鎏
+蜘
+蛛
+袂
+裆
+須
+覺
+遏
+瘋
+藝
+團
+睁
+蜊
+囧
+螃
+螨
+刍
+衍
+槟
+蚤
+芹
+闯
+誰
+萬
+歲
+诠
+诡
+谲
+吵
+脫
+庚
+嘟
+婺
+焯
+琛
+躬
+買
+劵
+旱
+吡
+啶
+臻
+婪
+绷
+卿
+澈
+胤
+锰
+沧
+級
+廚
+蛟
+躺
+嗝
+蓟
+饺
+贞
+寅
+賺
+哑
+匈
+弋
+轰
+衙
+罕
+豺
+惶
+侃
+嘈
+茗
+觑
+逛
+跪
+擒
+夷
+鱿
+邓
+讶
+邵
+彧
+懈
+楷
+僵
+溅
+馍
+钽
+锣
+羹
+酮
+翎
+製
+麥
+語
+匆
+躁
+盯
+俑
+呐
+隽
+擘
+睹
+歧
+竺
+瞟
+狰
+狞
+釜
+丐
+骐
+飲
+飽
+筆
+記
+飙
+嘎
+紅
+瑪
+咀
+铲
+杖
+馅
+鳄
+琪
+芪
+雌
+筷
+琶
+肇
+眶
+螳
+螂
+娓
+睬
+凛
+冽
+涌
+幢
+掐
+巿
+憋
+铿
+锵
+妄
+捲
+窠
+臼
+祗
+匪
+矗
+烩
+℉
+婧
+碾
+冉
+沼
+蕉
+燒
+抜
+込
+删
+薪
+抠
+铛
+Ⅰ
+伝
+聞
+穗
+蜥
+蜴
+襄
+邑
+窦
+壹
+潞
+舂
+課
+巳
+咒
+醛
+溉
+潍
+笃
+螟
+赘
+骼
+鳗
+脓
+鸽
+羧
+饶
+峪
+竽
+耽
+肘
+腈
+槐
+礎
+應
+孢
+聂
+卯
+榻
+菁
+沅
+穩
+氙
+禹
+缪
+孜
+櫃
+鹦
+鹉
+杵
+調
+順
+宮
+備
+羟
+茯
+Ｌ
+Ｃ
+Ｄ
+屎
+Ⅱ
+^
+樹
+髮
+茧
+簡
+娶
+惟
+鲸
+荆
+渲
+暑
+戮
+嗨
+祇
+塬
+髻
+〖
+関
+獲
+畫
+兮
+冯
+張
+圖
+擾
+弼
+歡
+無
+爭
+寧
+靜
+辊
+颔
+歐
+暹
+芋
+馥
+缕
+砗
+磲
+竭
+戾
+褚
+吏
+硫
+吝
+啬
+惩
+晗
+嬉
+帷
+琵
+蟑
+營
+號
+眾
+憩
+疽
+骇
+巷
+蘭
+誠
+報
+惘
+論
+蓓
+巫
+靡
+缦
+卻
+難
+堰
+骅
+荚
+鸢
+崭
+姥
+淬
+舶
+呜
+怯
+腼
+腆
+構
+祕
+钯
+莹
+溯
+淤
+钣
+倜
+傥
+骚
+衩
+濒
+嚼
+逮
+腥
+迂
+尸
+箐
+荫
+並
+揮
+盡
+從
+鄂
+苞
+拴
+扼
+喉
+礁
+闸
+眺
+绰
+黝
+邃
+琅
+氓
+娆
+茸
+蚧
+樟
+贏
+業
+廠
+冈
+盹
+枢
+洱
+愣
+亢
+炽
+睦
+瞒
+咆
+躲
+關
+涕
+迸
+噬
+呛
+瘪
+祀
+飨
+缮
+茵
+堕
+祠
+卞
+潇
+筠
+巍
+摒
+粕
+隶
+匮
+桨
+嚏
+葆
+眷
+驳
+丶
+荔
+僚
+萼
+箭
+酌
+榷
+雎
+缀
+焉
+霹
+雳
+敝
+捂
+剁
+倘
+費
+電
+璋
+韓
+蝽
+眩
+Ｇ
+凋
+钨
+谎
+捺
+撷
+叩
+幌
+挨
+叨
+Ｏ
+揩
+叽
+哩
+〈
+〉
+嘀
+嗒
+鳳
+莘
+昶
+闺
+哎
+`
+囤
+囔
+竖
+汞
+瘀
+喧
+哗
+痔
+肛
+焰
+怖
+藩
+彤
+瘿
+倫
+婷
+珐
+纭
+糅
+戌
+叼
+唏
+玺
+骊
+奎
+侈
+缇
+潦
+憔
+悴
+龚
+踝
+稼
+廖
+載
+銘
+漢
+妊
+娠
+妓
+婶
+俾
+豬
+攸
+署
+秸
+渺
+楦
+菖
+纰
+赌
+铋
+蹦
+懊
+蔼
+瞪
+拇
+菡
+嘛
+桓
+铠
+疳
+榫
+岫
+跻
+奄
+腓
+峨
+妍
+痘
+斥
+屑
+偕
+彪
+苷
+吁
+昇
+铢
+驯
+徙
+庶
+見
+馬
+內
+闽
+紋
+蜕
+鹃
+驷
+慑
+钵
+咙
+镣
+铐
+琼
+茜
+幔
+洮
+寵
+薦
+軍
+敞
+領
+峥
+嵘
+俎
+斡
+秃
+廣
+噻
+藥
+態
+蟆
+磊
+謊
+恺
+惦
+诧
+霾
+啮
+谔
+晔
+袄
+辍
+叱
+咤
+郝
+佯
+噶
+Ｑ
+鹫
+梓
+組
+賦
+親
+甾
+桖
+芷
+凄
+愫
+珏
+汝
+邺
+沁
+泫
+亩
+铒
+浣
+霓
+涿
+蔚
+嫔
+黔
+炊
+蒡
+筐
+馋
+捅
+玖
+柒
+哇
+灣
+鸳
+鸯
+砧
+痢
+沏
+涣
+馒
+骠
+畲
+確
+達
+戶
+柚
+啫
+谬
+踞
+廓
+較
+雖
+鬥
+颀
+塌
+睑
+颊
+睾
+夠
+傷
+潢
+結
+椿
+窈
+窕
+蹊
+漩
+鲈
+臧
+巽
+疱
+裘
+诃
+沌
+腴
+谶
+拱
+肱
+腱
+髂
+奚
+疟
+毽
+峭
+梢
+穹
+隍
+嗳
+渥
+柿
+缰
+嘌
+呤
+認
+敗
+毗
+酋
+捶
+艮
+珈
+芍
+芸
+茉
+薏
+蘸
+耸
+頁
+吭
+觊
+觎
+疙
+瘩
+蛊
+請
+針
+膚
+瑩
+觅
+摞
+寐
+遨
+嚣
+讥
+绯
+淌
+叟
+痤
+婢
+莺
+掷
+饵
+沮
+貴
+價
+猝
+岐
+識
+鳴
+Ｅ
+発
+麂
+舷
+輕
+舵
+辜
+艘
+瞰
+沱
+涎
+遁
+兀
+禀
+嫂
+芮
+笠
+還
+蜻
+蜓
+邱
+焘
+镙
+幡
+猗
+歹
+朔
+昔
+恣
+谑
+韬
+燮
+舫
+霏
+痊
+扦
+箕
+砼
+痨
+骶
+钒
+黃
+筏
+岔
+帛
+唬
+亳
+Ｘ
+膦
+劇
+塾
+苋
+甬
+扒
+吩
+咐
+碘
+苡
+吋
+熵
+驅
+陰
+磺
+锛
+枳
+浔
+麿
+誘
+連
+舊
+擋
+咔
+唑
+鼹
+肴
+桅
+秧
+損
+潔
+劑
+钼
+芩
+湯
+麵
+垄
+漱
+門
+網
+狀
+⑧
+█
+幫
+碩
+紊
+隱
+搐
+斟
+咎
+荅
+皋
+拈
+評
+爲
+臺
+農
+產
+標
+丞
+茴
+蔺
+蝗
+肆
+虐
+墩
+靳
+袤
+崂
+豌
+甄
+砾
+獭
+勐
+懿
+铵
+華
+慨
+宸
+壑
+磋
+讫
+僅
+釐
+鳕
+僻
+镑
+恬
+栀
+逍
+薹
+汕
+膠
+昱
+獐
+膺
+侏
+黯
+荞
+啼
+诰
+嫖
+陡
+俚
+亘
+篡
+莼
+砥
+汶
+佘
+夥
+宕
+谴
+牒
+洼
+霭
+阉
+镭
+俐
+貧
+阖
+傣
+豉
+吒
+鋼
+線
+條
+膩
+锢
+郵
+彷
+徨
+沖
+胛
+仨
+锗
+汹
+谖
+荃
+飓
+卅
+飒
+恍
+惚
+郡
+鞠
+攘
+邸
+黜
+桎
+梏
+牍
+減
+慾
+嗦
+弩
+铆
+蹿
+則
+禮
+哔
+崤
+铰
+﹑
+蚌
+惡
+旷
+匕
+堅
+強
+祉
+咫
+夯
+舱
+稔
+睽
+淄
+膑
+獾
+夭
+簌
+蝌
+蚪
+吮
+煨
+殇
+蜉
+蝣
+鋪
+逗
+佗
+掬
+慷
+徕
+赦
+熄
+愠
+嬷
+撸
+煦
+杲
+窒
+莽
+泯
+紛
+籠
+腩
+蝈
+斛
+驮
+阡
+遐
+∥
+苇
+熹
+胭
+嚎
+毓
+氰
+鞅
+梧
+槁
+呻
+雜
+誌
+蕙
+瞳
+芘
+痞
+龋
+琏
+焖
+牦
+鹩
+祟
+迥
+簪
+玥
+歼
+瓯
+蘖
+茁
+痉
+垩
+喳
+飛
+稷
+訊
+酉
+恆
+讀
+鸾
+赊
+弑
+壅
+绦
+祷
+魇
+哨
+粟
+嶙
+峋
+渝
+伉
+涧
+懦
+瘸
+蕻
+煜
+楫
+豐
+聯
+絡
+謝
+赅
+钿
+螈
+虛
+攝
+補
+蘇
+夔
+舔
+纂
+樽
+赝
+蕃
+霁
+蹇
+傀
+儡
+烊
+擞
+冢
+赡
+悅
+窘
+陇
+枋
+鲣
+笆
+汜
+篱
+雲
+闳
+骡
+鹑
+復
+龅
+繫
+呕
+懋
+區
+譽
+寫
+擢
+簇
+戎
+麸
+腧
+東
+揍
+暨
+唧
+傑
+晉
+暢
+銷
+Ｓ
+椽
+況
+咄
+厩
+湍
+蛭
+楔
+碁
+擬
+彙
+螢
+蟲
+笈
+霄
+骰
+呦
+吳
+單
+顆
+鎖
+缉
+岖
+壕
+垣
+眯
+虢
+鹌
+遊
+戲
+Ｕ
+吱
+椴
+泠
+錾
+財
+榛
+圃
+蜈
+蚣
+鹭
+蜷
+蜍
+笞
+弭
+缥
+缈
+婕
+佼
+揖
+拮
+茭
+臆
+恻
+镌
+紹
+瞥
+嚷
+倏
+皎
+俨
+維
+護
+蘅
+荪
+喃
+鲲
+炳
+晤
+剽
+犷
+啾
+鏖
+囿
+屌
+譬
+貉
+寮
+蛹
+唛
+腫
+熱
+伢
+掸
+輔
+導
+響
+殼
+娲
+撇
+寰
+酗
+雉
+瘙
+猬
+囡
+掏
+乍
+彗
+讴
+菏
+淹
+項
+産
+婦
+啰
+羁
+咿
+唷
+覽
+鹊
+孑
+聲
+捌
+摧
+納
+隨
+饷
+‖
+犸
+谧
+朦
+胧
+飾
+岀
+蜿
+蜒
+荏
+殒
+嗓
+訂
+詢
+瞅
+莅
+譯
+驟
+檢
+診
+挠
+紙
+屿
+涟
+漪
+別
+鳍
+铱
+迩
+蓑
+濮
+瘠
+咛
+煩
+惱
+宓
+氤
+氲
+荻
+唢
+笺
+煅
+鲢
+鳙
+噱
+龈
+禾
+贲
+躍
+犁
+膈
+谒
+狈
+镗
+奧
+骈
+衄
+悯
+姊
+骥
+貼
+緒
+蕲
+侮
+垠
+肋
+嵴
+緩
+壓
+渚
+阕
+瓤
+倭
+銀
+變
+橫
+瓠
+篝
+谤
+隘
+缭
+轉
+參
+镊
+係
+糠
+钕
+璟
+锲
+盤
+Ｆ
+莨
+灏
+赓
+鹧
+鸪
+
+糜
+廳
+呸
+裏
+絕
+坂
+暧
+脘
+峦
+崽
+帳
+稅
+啉
+叢
+闰
+Ⅶ
+涝
+稗
+試
+験
+痈
+熠
+捩
+搂
+宵
+恙
+該
+織
+⑥
+鲇
+螯
+钡
+範
+邊
+畢
+鳌
+暮
+铎
+館
+捍
+劍
+癜
+殺
+仟
+诅
+監
+楽
+浚
+媳
+棣
+垦
+痣
+楣
+倔
+跷
+匾
+喰
+鬓
+週
+衝
+聖
+綜
+犊
+忱
+骸
+憎
+忒
+屹
+谙
+祐
+笙
+澍
+敖
+庖
+咋
+蒹
+葭
+坳
+蔑
+麗
+痪
+耙
+礙
+淨
+島
+韶
+捎
+邬
+蛔
+疥
+癞
+嫣
+翦
+啪
+嚓
+绗
+搀
+執
+龍
+兒
+踱
+岌
+撵
+俺
+锄
+诂
+擺
+芾
+帶
+慶
+筵
+棺
+鍵
+濠
+珩
+徜
+徉
+蛰
+惭
+習
+掛
+濑
+嘗
+漕
+阪
+磐
+懑
+糸
+馊
+窿
+癖
+蚜
+榔
+蓖
+阮
+丨
+锶
+蟠
+锭
+湄
+滁
+藓
+淼
+斫
+撩
+嬛
+髁
+癎
+贬
+鬆
+菸
+瓮
+痿
+謀
+鄙
+忏
+椹
+芜
+姗
+髯
+虬
+赭
+頫
+孺
+丕
+階
+擇
+郴
+摺
+蟒
+滢
+員
+狒
+绉
+纣
+篓
+﹝
+﹞
+譜
+悌
+Ｉ
+Ｂ
+Ｎ
+Ｍ
+袒
+攫
+茼
+偉
+唠
+劃
+臊
+箝
+瞌
+疣
+糧
+隻
+砷
+嗲
+證
+Ⅺ
+螭
+顏
+惋
+溝
+兢
+粼
+昉
+陉
+泞
+紗
+預
+収
+冊
+齊
+膻
+辕
+痼
+琬
+猾
+峙
+Ｐ
+礴
+畿
+纨
+勞
+厮
+禺
+轶
+淦
+幾
+皲
+谕
+楝
+蛆
+澎
+湃
+墉
+玮
+雹
+蚬
+胄
+锹
+陂
+剐
+鄞
+跡
+眈
+麓
+遛
+穰
+苴
+協
+扈
+缢
+蛏
+鲟
+蛎
+瞠
+帚
+臉
+蟋
+蟀
+郸
+驿
+氐
+襁
+褓
+驸
+赈
+锺
+镲
+孰
+韫
+蠕
+忻
+踹
+唆
+恸
+汆
+腮
+烏
+瀛
+訓
+練
+飕
+栢
+榈
+陨
+殚
+養
+濕
+欖
+馀
+茹
+鈴
+溟
+饴
+吲
+哚
+賞
+藍
+顷
+茱
+蓮
+枷
+佃
+盥
+燙
+雙
+換
+诋
+鬃
+犒
+铨
+缨
+嫡
+擂
+楊
+蘿
+蔔
+遴
+飚
+邋
+遢
+跖
+聩
+袅
+喙
+闫
+嗮
+铳
+冼
+純
+炯
+負
+責
+憶
+礦
+餃
+˙
+鳅
+氽
+旎
+谟
+議
+顯
+權
+嬴
+纾
+玑
+複
+聰
+闾
+熘
+奂
+哧
+皈
+璜
+倬
+錢
+悖
+翊
+苻
+饯
+钏
+廟
+輸
+祜
+鲂
+镉
+亟
+犟
+貓
+邰
+娣
+側
+浜
+簽
+Ⅹ
+喆
+註
+釋
+噼
+诽
+袱
+胯
+绥
+臃
+伎
+匝
+攥
+钜
+樾
+興
+遞
+愕
+鎮
+搽
+徳
+綱
+榉
+踌
+躇
+桉
+苄
+膽
+姣
+吆
+沓
+铌
+獒
+畦
+縫
+牠
+滾
+阆
+渭
+Ⅴ
+廪
+嚴
+侶
+窮
+帼
+羅
+聽
+懵
+臨
+蝰
+蜱
+蠡
+褒
+嶂
+莊
+汐
+裨
+钎
+晷
+燧
+Ⅸ
+菀
+鳜
+荀
+珅
+蚓
+繡
+肮
+査
+檔
+講
+崴
+実
+坍
+Ⅷ
+萸
+陳
+▽
+閱
+欄
+睢
+撂
+邕
+颞
+醬
+酊
+侥
+甥
+蠹
+骛
+蹒
+跚
+鞣
+綁
+呗
+瘘
+痱
+孛
+貝
+跛
+韻
+岑
+麝
+腸
+桿
+鳥
+捨
+棄
+擴
+矜
+揚
+孪
+姒
+轱
+辘
+枇
+杷
+柘
+绡
+轫
+Ｋ
+Ａ
+Ｒ
+歷
+輝
+獻
+橋
+鹳
+捭
+￠
+淚
+澂
+晁
+苜
+蓿
+闩
+桁
+耷
+︱
+烬
+販
+颛
+鹂
+漉
+晌
+鬣
+醪
+馕
+禍
+帙
+願
+鹘
+煽
+Ｙ
+Ｚ
+颧
+怦
+髌
+榧
+檫
+妝
+勝
+铑
+⑨
+鮮
+燉
+啜
+啕
+瓒
+绮
+芫
+荽
+檸
+钺
+汛
+﹒
+皑
+栉
+嬗
+苫
+閒
+潛
+珞
+娄
+炕
+坨
+潼
+缜
+劉
+〞
+锱
+馁
+鹗
+刈
+凫
+髎
+颢
+謇
+逶
+迤
+舉
+酐
+軟
+涪
+偈
+臟
+蹩
+猷
+篦
+锴
+娉
+衅
+賓
+蚯
+蝮
+擔
+職
+媽
+铮
+鐘
+齡
+猄
+谡
+黍
+‐
+萘
+煸
+鍋
+矽
+鉻
+⑦
+桧
+陛
+俳
+荥
+幄
+呱
+勢
+葉
+邳
+聃
+繇
+綿
+徹
+罡
+跺
+嵊
+霎
+閉
+裾
+諸
+隊
+≧
+≦
+搖
+哕
+緣
+讷
+龌
+龊
+藿
+匍
+匐
+埚
+諾
+瞭
+涸
+芈
+姝
+缙
+颍
+莜
+衢
+诬
+葚
+濃
+廢
+肫
+慣
+鲛
+
+芎
+萜
+脲
+掳
+鐵
+殴
+蓼
+潴
+箩
+屆
+獎
+剛
+頻
+掖
+旖
+輩
+煊
+擀
+妤
+蜣
+濟
+漸
+哌
+顛
+榮
+挈
+魯
+蹂
+躏
+诤
+饋
+掣
+嘹
+咧
+涼
+鄉
+夢
+燎
+怆
+拚
+椁
+嗔
+輪
+涔
+锷
+踐
+陣
+禳
+鄱
+衾
+鍛
+鍊
+廿
+栎
+诟
+臌
+疖
+藐
+粽
+賀
+芡
+觸
+柞
+褂
+哞
+餅
+絲
+恽
+俸
+層
+揪
+茲
+潑
+阂
+貢
+詩
+垅
+碉
+壞
+骞
+骁
+珉
+単
+樨
+鞘
+罔
+呃
+囫
+陲
+蒯
+擊
+鏈
+婿
+対
+攪
+傾
+琨
+嗎
+戍
+橇
+馗
+赳
+崧
+诌
+偎
+麋
+舸
+逋
+嗖
+埭
+贈
+嗪
+徑
+偌
+腳
+虻
+佞
+谗
+佝
+偻
+審
+皴
+潸
+杈
+鄢
+碴
+檄
+婵
+唁
+瘁
+杳
+缄
+闆
+迴
+楹
+镛
+誕
+魃
+頂
+澧
+炝
+遷
+濡
+爻
+旌
+捋
+舐
+墊
+蓦
+颓
+縱
+劬
+薮
+郗
+≮
+≯
+訴
+築
+荼
+伫
+歆
+邈
+緊
+嘚
+詮
+嗤
+吖
+甙
+蛲
+獗
+蛐
+駕
+仞
+覃
+锕
+円
+閃
+樓
+罂
+様
+醴
+谆
+圄
+旸
+蚩
+埙
+徭
+喬
+塊
+煙
+椐
+諧
+喹
+聿
+酞
+硐
+粲
+骧
+瓴
+雞
+錯
+幣
+牆
+飯
+唰
+億
+抨
+噌
+險
+羿
+遒
+塗
+幹
+抡
+耆
+搔
+黒
+嗷
+薩
+娼
+醺
+勻
+讚
+宥
+彿
+谄
+髙
+Ｊ
+楮
+胗
+桀
+褛
+颚
+壽
+蔥
+啧
+п
+я
+咂
+轲
+迢
+琮
+額
+舛
+颏
+溏
+囟
+纖
+鏡
+獠
+抻
+谩
+鼬
+淖
+鲆
+啖
+炷
+槌
+劾
+蓋
+樑
+貫
+訪
+競
+鹹
+攜
+沢
+瘰
+疬
+笫
+觐
+説
+郦
+苎
+穀
+遺
+駐
+啟
+鴉
+壯
+槭
+琰
+霈
+蓍
+臬
+儀
+趨
+暫
+蘋
+蕹
+祂
+簋
+黠
+葳
+邁
+咝
+惺
+忡
+菅
+蒌
+薤
+鍾
+嘭
+怏
+羯
+瞽
+醍
+靛
+祚
+窖
+焱
+綠
+糾
+莳
+顼
+仝
+岷
+猖
+鲷
+諮
+顔
+蘊
+憑
+醌
+嗡
+鲶
+穑
+塵
+撫
+谪
+陸
+钤
+耄
+耋
+鲵
+幺
+苒
+蚵
+餡
+涞
+◣
+◢
+鑑
+陟
+楸
+踉
+跄
+帏
+Ｗ
+拄
+蝦
+騎
+栝
+箧
+挲
+燊
+绾
+剋
+阗
+瘕
+祿
+夙
+蒐
+泱
+Ｈ
+饬
+獅
+鲽
+猞
+猁
+呂
+蕅
+椤
+懼
+庾
+闱
+沣
+辭
+違
+齒
+鹽
+紐
+峒
+琊
+賴
+卟
+癔
+弁
+掇
+谌
+偃
+醣
+鲳
+褴
+杼
+災
+専
+砜
+遜
+辺
+唎
+呓
+掂
+倉
+縣
+梆
+勁
+锑
+気
+嗟
+証
+唔
+銅
+裥
+蔫
+濂
+呋
+◤
+傢
+囍
+萊
+爛
+曬
+绻
+胝
+>
+·
+昀
+°
+&
+←
+↓
+縻
+"
+”
+騄
+“
+≤
+≥
+<
+馝
+²
+褯
+√
+淜
+凍
+幚
+綫
+觇
+醮
+艏
+鲗
+脟
+³
+蒱
+罽
+鈣
+豊
+愎
+柅
+妯
+膘
+¥
+钍
+尕
+虓
+凈
+‘
+’
+噔
+頌
+±
+醨
+毵
+崋
+絨
+玃
+葖
+…
+—
+菪
+䏡
+陬
+檞
+洇
+汦
+缬
+扱
+瀣
+祋
+Ω
+尻
+骉
+μ
+→
+殁
+腽
+燚
+衮
+粧
+鷉
+薬
+≈
+冁
+騵
+謏
+乜
+傘
+勖
+勲
+嚭
+砹
+夐
+郜
+黼
+䗛
+嵗
+鄀
+牖
+㭎
+▪
+豳
+▶
+樋
+鳃
+◀
+黇
+輋
+麽
+葎
+砵
+穄
+甭
+綄
+洌
+芤
+硚
+酹
+籤
+﹣
+×
+俍
+鏏
+↑
+鵏
+眞
+嬂
+≠
+盜
+緯
+縝
+溆
+亓
+騠
+砬
+鐍
+‥
+栊
+翮
+鲘
+藟
+墖
+棐
+粝
+迺
+趼
+吣
+庠
+鮭
+纇
+満
+骺
+睨
+黉
+脞
+媄
+胨
+郓
+濋
+圜
+繄
+锫
+滆
+澫
+潏
+罘
+譓
+叻
+殳
+癢
+哐
+皛
+偼
+诮
+鼽
+䕶
+窨
+䡵
+疁
+遘
+钚
+卺
+肟
+擿
+鹼
+鄗
+彳
+醯
+雱
+貯
+鮑
+坻
+缁
+瓀
+噁
+髪
+鳯
+煳
+嵫
+伈
+窀
+堃
+軒
+豕
+礌
+晳
+•
+锏
+骦
+晡
+嬿
+朸
+棪
+铧
+勰
+缧
+囵
+駓
+迕
+↘
+↖
+岡
+镩
+齉
+氡
+尨
+逑
+鰤
+珮
+怃
+譲
+婤
+譩
+譆
+膂
+滧
+踺
+∞
+耢
+豨
+勠
+孅
+枅
+殄
+馭
+噠
+鄑
+賜
+嬰
+鈍
+鈿
+臍
+噙
+朐
+苤
+垈
+嬄
+饧
+雠
+菰
+辂
+嵽
+甍
+榅
+閏
+繃
+蘆
+薈
+汙
+腒
+冮
+佽
+黾
+隺
+脬
+铏
+€
+梠
+鏽
+噴
+霧
+崶
+脅
+餾
+絜
+佸
+勵
+鮈
+蔄
+畤
+缐
+洓
+躅
+绀
+㛹
+诶
+犄
+汋
+忖
+慝
+⑩
+枼
+垕
+搌
+翂
+铷
+阄
+菓
+廛
+肓
+龠
+滪
+劓
+玟
+缛
+饳
+嵛
+∧
+蟎
+虒
+絷
+龆
+鲰
+翾
+秕
+纕
+镬
+戕
+寳
+抿
+舖
+詈
+繩
+牽
+艹
+纆
+儉
+愍
+躞
+娵
+镪
+羕
+龉
+镝
+罴
+硌
+輗
+菂
+犇
+暲
+鑒
+輦
+镵
+締
+儲
+䲢
+诹
+軀
+喚
+ˉ
+扂
+篢
+贜
+ω
+姘
+烔
+爰
+圉
+↙
+↗
+彄
+跬
+傒
+釿
+袢
+跗
+諟
+绐
+骜
+▏
+腚
+旐
+廋
+橦
+隹
+囹
+雊
+邙
+绖
+戣
+刖
+筻
+￡
+綎
+赗
+氍
+舄
+檑
+鼗
+償
+遲
+閣
+囲
+鲊
+潵
+甌
+慥
+瘛
+荇
+鐽
+▍
+∣
+悃
+蹐
+鹾
+鹜
+娈
+蹉
+扺
+欹
+貘
+碓
+锖
+薁
+鼱
+锍
+杓
+兿
+惝
+厖
+劼
+橛
+稂
+酅
+猃
+芼
+糒
+硖
+磡
+憕
+弨
+脔
+龃
+揕
+嵖
+﹢
+旆
+镨
+骝
+褊
+郚
+歅
+笪
+纛
+牥
+薳
+镞
+荬
+锧
+诜
+瀼
+孳
+斠
+圊
+骓
+︿
+袪
+雒
+趯
+珸
+捃
+﹥
+秬
+嫒
+蹁
+跹
+鹛
+吥
+鼍
+舯
+卬
+淙
+羼
+酽
+嘞
+嘧
+苾
+諓
+眄
+騊
+颙
+鄚
+蘘
+潋
+滟
+鳊
+觭
+禋
+闉
+盍
+哳
+蚰
+骟
+裣
+扃
+宄
+嫚
+袼
+蹙
+栲
+∈
+鐡
+饅
+嗄
+郿
+炣
+塰
+蛳
+熥
+瑄
+枞
+俢
+芨
+妏
+埕
+晞
+嬅
+媞
+萩
+倮
+価
+瀾
+琯
+珺
+暘
+偮
+玎
+襪
+璎
+祎
+驲
+蔪
+α
+晢
+嬤
+屛
+姹
+銥
+璽
+妡
+転
+咲
+柰
+苘
+牪
+洧
+绋
+澋
+鷄
+盱
+眙
+袆
+鈷
+䂀
+苕
+旻
+珰
+轳
+訫
+咭
+莵
+噗
+浠
+ˆ
+婊
+湓
+嗑
+孬
+倆
+袴
+瀘
+饽
+▫
+芃
+瀞
+熳
+蒼
+煖
+杺
+咻
+㧴
+壇
+嵮
+桤
+兲
+灆
+塢
+铪
+瘊
+孖
+萆
+薢
+蔾
+凔
+棧
+葐
+﹩
+哓
+哂
+柫
+缟
+垚
+埗
+洹
+厝
+玙
+衿
+佲
+槎
+岙
+峽
+綺
+謎
+蒄
+曈
+塍
+荳
+佧
+砀
+刄
+駿
+蓢
+圩
+捘
+阊
+悻
+狍
+郞
+偲
+伲
+淏
+唻
+臜
+塱
+郫
+浉
+睏
+疃
+笕
+畈
+弇
+縁
+茌
+憇
+紡
+昄
+啵
+浐
+肈
+谞
+俹
+沩
+柁
+俫
+镆
+垟
+洺
+箬
+濛
+鏌
+茡
+苪
+瑷
+団
+茆
+槙
+蕗
+糗
+邝
+褟
+铉
+鲱
+俣
+滐
+咮
+蠟
+鲅
+沭
+鲋
+漯
+蒽
+灥
+镫
+峄
+湑
+俁
+宬
+酩
+珲
+铖
+塚
+戦
+郯
+瑀
+怂
+禤
+錡
+蔲
+锜
+煕
+碶
+蘼
+赑
+繆
+僖
+莒
+㚢
+菘
+鸲
+渃
+β
+扪
+谵
+橼
+螫
+蕈
+猥
+亵
+疋
+蚴
+眦
+襞
+胼
+绌
+擤
+腠
+纡
+胬
+蹼
+霰
+襻
+榼
+蒺
+岢
+脒
+莪
+葶
+莶
+巯
+胍
+菴
+醑
+癀
+檗
+肼
+癃
+癥
+菥
+蓂
+蝥
+酏
+鳢
+泮
+鉍
+荜
+茇
+菝
+葜
+薷
+蟅
+尪
+腙
+芴
+蒎
+棓
+楤
+呫
+Ι
+γ
+胂
+蠲
+茚
+镧
+礞
+艽
+艿
+苈
+蓣
+钆
+鳔
+鸬
+鹚
+鸶
+φ
+δ
+缗
+铯
+羸
+鉧
+毳
+啘
+喑
+′
+噤
+噫
+耵
+聍
+矇
+搦
+忤
+瘅
+瘵
+疰
+痃
+瘴
+蛉
+劄
+睥
+瞀
+羰
+聤
+廔
+腘
+臁
+蓐
+铊
+蟥
+黧
+銹
+澆
+鋁
+錫
+閑
+絢
+搶
+鄭
+擲
+遙
+鈔
+鑽
+濾
+錦
+貿
+輻
+盞
+渦
+攤
+灘
+績
+嘆
+訟
+脹
+帥
+掃
+鈎
+闊
+褲
+襯
+屜
+櫥
+慮
+賠
+擠
+騰
+瀝
+賢
+瓊
+悶
+脈
+箏
+皺
+奮
+槍
+爐
+嬌
+韌
+瞞
+駛
+膿
+瘡
+滌
+氫
+憂
+釀
+誼
+疊
+鴨
+蠶
+鵝
+罵
+睞
+噸
+鞏
+嗆
+緬
+攏
+謙
+縛
+繞
+蕩
+鑰
+罰
+叄
+貳
+崗
+軌
+銳
+趙
+滬
+剝
+鄧
+諒
+濺
+繽
+紉
+賃
+鑲
+頑
+氈
+艱
+釣
+絞
+綉
+瀏
+頸
+撲
+殘
+腎
+賬
+厭
+鶴
+獸
+暈
+橢
+寢
+翹
+謹
+鴻
+艦
+墜
+穎
+懶
+敵
+朧
+鈉
+馳
+謠
+嶺
+饒
+喲
+啞
+瀉
+瑣
+蔣
+窩
+紳
+墳
+蝸
+螞
+蟻
+豎
+鈕
+撥
+賈
+濁
+襲
+俠
+煥
+鉛
+艷
+黨
+錐
+彎
+煉
+憐
+毆
+鬧
+諱
+毀
+滅
+鈞
+獄
+匯
+陝
+馮
+闖
+趕
+龜
+顫
+濱
+曉
+蠅
+漿
+輿
+簾
+喪
+鍬
+燦
+債
+憲
+棗
+鋅
+辯
+鋒
+虧
+釘
+窯
+飄
+廈
+熒
+錘
+磚
+閥
+槳
+贊
+鑄
+懸
+臘
+遼
+闡
+樺
+綴
+鍍
+滯
+貸
+矯
+肅
+漲
+壺
+鯽
+漁
+厲
+偵
+邏
+蝕
+攔
+摳
+禱
+誨
+飼
+淵
+騙
+滲
+閨
+凜
+鉀
+潰
+綽
+齋
+聳
+轎
+蕭
+癬
+纜
+狹
+錳
+鷹
+彌
+爺
+鯊
+廂
+籃
+偽
+頒
+綳
+誦
+禪
+鐺
+娛
+滷
+鱗
+龐
+墾
+蠻
+灑
+墮
+窺
+鄰
+嬸
+囑
+籌
+轟
+緞
+侖
+嶼
+頗
+鉗
+驕
+屢
+騷
+憤
+摻
+豈
+硯
+絹
+燭
+瀟
+盧
+爍
+摯
+憫
+斂
+顱
+縷
+轄
+贅
+貞
+雛
+嚇
+賊
+蕪
+濤
+輛
+殲
+犧
+疇
+嚨
+攆
+隸
+貪
+詠
+僑
+拋
+篩
+鏟
+驢
+檻
+綻
+滄
+癮
+諷
+辮
+獰
+葦
+饑
+餓
+澀
+竊
+頰
+餌
+尷
+奪
+繭
+鴿
+淪
+粵
+掙
+覓
+簫
+閩
+駱
+駝
+鵬
+憊
+瘧
+擰
+嶄
+楓
+勸
+詭
+棟
+賤
+謬
+睜
+繹
+銜
+嘯
+閘
+濫
+攬
+鴕
+綢
+撿
+樞
+艙
+壟
+隕
+蔭
+諜
+饞
+欽
+擱
+嘔
+譚
+慘
+鯨
+懇
+懲
+誇
+鋸
+蟬
+罷
+鰐
+棲
+斃
+鐮
+莖
+巔
+馱
+撈
+狽
+鸚
+鵡
+糞
+幟
+渾
+摟
+畝
+詐
+鵑
+諺
+壘
+樁
+葷
+譏
+澗
+纏
+誡
+嗚
+駁
+駭
+貽
+鴛
+鴦
+囂
+鯉
+藹
+輾
+繳
+襖
+賭
+嘩
+錨
+婁
+譴
+鵲
+曠
+駒
+籮
+竄
+靂
+贛
+撓
+斬
+廬
+僥
+堯
+饃
+鑼
+頃
+搗
+壩
+鶯
+勛
+鱉
+軋
+鑿
+轍
+釁
+賄
+簍
+騾
+聶
+訝
+矚
+頹
+貶
+閻
+鷗
+繮
+馴
+鎊
+蹺
+慚
+礫
+嘰
+絆
+鬢
+癟
+礬
+聾
+鰭
+繚
+鋤
+猙
+濘
+巒
+癱
+晝
+嘮
+籬
+謗
+緝
+穢
+虜
+諄
+稟
+瘓
+瀕
+渙
+洶
+贖
+揀
+嶇
+鎬
+骯
+醞
+攙
+澇
+餒
+銬
+贓
+挾
+訛
+賂
+誣
+窪
+邗
+滦
+埇
+郧
+浈
+堽
+垸
+珙
+邛
+崃
+岘
+垭
+濉
+牯
+稜
+碚
+谯
+浯
+兖
+凇
+芗
+耒
+灞
+渑
+渌
+甪
+邡
+滘
+浍
+潆
+傈
+僳
+磴
+仵
+泺
+紶
+葑
+藁
+迳
+垌
+泷
+矸
+墒
+篁
+汊
+幛
+鄄
+漖
+秭
+儋
+湟
+枧
+厍
+崮
+砦
+犍
+鄯
+赉
+郏
+秣
+隰
+尓
+垱
+蓥
+呷
+蕰
+坭
+汨
+瀍
+沔
+垻
+畹
+蚶
+鮀
+辇
+沚
+垴
+垡
+淅
+埯
+剡
+桷
+埒
+邨
+圪
+夼
+仡
+猇
+堍
+洸
+溱
+胪
+湫
+洄
+笏
+辋
+郢
+亍
+崆
+泖
+珥
+黟
+濯
+昝
+滏
+洎
+埝
+窢
+褔
+沤
+峤
+沨
+乂
+餍
+铦
+嵬
+赟
+訾
+墁
+锟
+逄
+肸
+姵
+劭
+璁
+酆
+贠
+玢
+邴
+弢
+臾
+薽
+牮
+戢
+靑
+曌
+隗
+翀
+洨
+鋆
+倧
+鍏
+璠
+贇
+钫
+赇
+珵
+錟
+炅
+頋
+珣
+粦
+尡
+璩
+婳
+甑
+篤
+镔
+嶷
+洐
+焜
+樯
+舲
+媗
+竑
+嫄
+澣
+甦
+麴
+娚
+琎
+韡
+钇
+謤
+骢
+穸
+眭
+唅
+鄷
+畇
+啹
+峯
+汔
+妎
+婂
+萏
+秝
+芄
+叡
+赜
+湉
+缑
+敻
+娒
+嘨
+礽
+缊
+
+傛
+阚
+澹
+昳
+琚
+翥
+翚
+昦
+逯
+飏
+琤
+勍
+钭
+椋
+湧
+囝
+翃
+镱
+堉
+庹
+瑗
+钲
+怿
+劢
+镄
+岍
+琍
+崟
+鹈
+ā
+á
+ǎ
+à
+ē
+é
+ě
+è
+ī
+í
+ǐ
+ì
+ō
+ó
+ǒ
+ò
+ū
+ú
+ǔ
+ù
+‰
+ 
+¢
+£
+฿
+
+

checkpoints/net_prior_860000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca229443cdbbd3dbe0fbaed2ffb77d5781473fa65bd8286408546a389e44e48c
+size 111897207

checkpoints/net_sr_860000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f06c9d7e195c529015fee9ef5f14f7b501ebeaba42f8baecdd86694bafd00841
+size 67812621

checkpoints/net_w_encoder_860000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c2626a4d3998b3edd88c60c2cdada721ea7531b366484c20a874819cbb426e5
+size 87585199