diff --git a/modules/image/Image_gan/style_transfer/face_parse/README.md b/modules/image/Image_gan/style_transfer/face_parse/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..a894d55f5c4a23fd6b29d2af9d1cdfd333a6c15e
--- /dev/null
+++ b/modules/image/Image_gan/style_transfer/face_parse/README.md
@@ -0,0 +1,133 @@
+# face_parse
+
+|模型名称|face_parse|
+| :--- | :---: |
+|类别|图像 - 人脸解析|
+|网络|BiSeNet|
+|数据集|-|
+|是否支持Fine-tuning|否|
+|模型大小|77MB|
+|最新更新日期|2021-12-07|
+|数据指标|-|
+
+
+## 一、模型基本信息
+
+- ### 应用效果展示
+ - 样例结果示例：
+
+ 
+
+ 输入图像
+
+ 
+
+ 输出图像
+
+
+
+- ### 模型介绍
+
+ - 人脸解析是语义图像分割的一种特殊情况，人脸解析是计算人脸图像中不同语义成分(如头发、嘴唇、鼻子、眼睛等)的像素级标签映射。给定一个输入的人脸图像，人脸解析将为每个语义成分分配一个像素级标签。
+
+
+
+## 二、安装
+
+- ### 1、环境依赖
+ - ppgan
+ - dlib
+
+- ### 2、安装
+
+ - ```shell
+ $ hub install face_parse
+ ```
+ - 如您安装时遇到问题，可参考：[零基础windows安装](../../../../docs/docs_ch/get_start/windows_quickstart.md)
+ | [零基础Linux安装](../../../../docs/docs_ch/get_start/linux_quickstart.md) | [零基础MacOS安装](../../../../docs/docs_ch/get_start/mac_quickstart.md)
+
+## 三、模型API预测
+
+- ### 1、命令行预测
+
+ - ```shell
+ # Read from a file
+ $ hub run face_parse --input_path "/PATH/TO/IMAGE"
+ ```
+ - 通过命令行方式实现人脸解析模型的调用，更多请见 [PaddleHub命令行指令](../../../../docs/docs_ch/tutorial/cmd_usage.rst)
+
+- ### 2、预测代码示例
+
+ - ```python
+ import paddlehub as hub
+
+ module = hub.Module(name="face_parse")
+ input_path = ["/PATH/TO/IMAGE"]
+ # Read from a file
+ module.style_transfer(paths=input_path, output_dir='./transfer_result/', use_gpu=True)
+ ```
+
+- ### 3、API
+
+ - ```python
+ style_transfer(images=None, paths=None, output_dir='./transfer_result/', use_gpu=False, visualization=True):
+ ```
+ - 人脸解析转换API。
+
+ - **参数**
+
+ - images (list\[numpy.ndarray\]): 图片数据，ndarray.shape 为 \[H, W, C\]；
+ - paths (list\[str\]): 图片的路径；
+ - output\_dir (str): 结果保存的路径；
+ - use\_gpu (bool): 是否使用 GPU；
+ - visualization(bool): 是否保存结果到本地文件夹
+
+
+## 四、服务部署
+
+- PaddleHub Serving可以部署一个在线人脸解析转换服务。
+
+- ### 第一步：启动PaddleHub Serving
+
+ - 运行启动命令：
+ - ```shell
+ $ hub serving start -m face_parse
+ ```
+
+ - 这样就完成了一个人脸解析转换的在线服务API的部署，默认端口号为8866。
+
+ - **NOTE:** 如使用GPU预测，则需要在启动服务之前，请设置CUDA\_VISIBLE\_DEVICES环境变量，否则不用设置。
+
+- ### 第二步：发送预测请求
+
+ - 配置好服务端，以下数行代码即可实现发送预测请求，获取预测结果
+
+ - ```python
+ import requests
+ import json
+ import cv2
+ import base64
+
+
+ def cv2_to_base64(image):
+ data = cv2.imencode('.jpg', image)[1]
+ return base64.b64encode(data.tostring()).decode('utf8')
+
+ # 发送HTTP请求
+ data = {'images':[cv2_to_base64(cv2.imread("/PATH/TO/IMAGE"))]}
+ headers = {"Content-type": "application/json"}
+ url = "http://127.0.0.1:8866/predict/face_parse"
+ r = requests.post(url=url, headers=headers, data=json.dumps(data))
+
+ # 打印预测结果
+ print(r.json()["results"])
+
+## 五、更新历史
+
+* 1.0.0
+
+ 初始发布
+
+ - ```shell
+ $ hub install face_parse==1.0.0
+ ```
diff --git a/modules/image/Image_gan/style_transfer/face_parse/model.py b/modules/image/Image_gan/style_transfer/face_parse/model.py
new file mode 100644
index 0000000000000000000000000000000000000000..c5df633416cd0ddc199bbb4bc7908e9dec008c58
--- /dev/null
+++ b/modules/image/Image_gan/style_transfer/face_parse/model.py
@@ -0,0 +1,51 @@
+# copyright (c) 2020 PaddlePaddle Authors. All Rights Reserve.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import sys
+import argparse
+
+from PIL import Image
+import numpy as np
+import cv2
+
+import ppgan.faceutils as futils
+from ppgan.utils.preprocess import *
+from ppgan.utils.visual import mask2image
+
+
+class FaceParsePredictor:
+ def __init__(self):
+ self.input_size = (512, 512)
+ self.up_ratio = 0.6 / 0.85
+ self.down_ratio = 0.2 / 0.85
+ self.width_ratio = 0.2 / 0.85
+ self.face_parser = futils.mask.FaceParser()
+
+ def run(self, image):
+ image = Image.fromarray(image)
+ face = futils.dlib.detect(image)
+
+ if not face:
+ return
+ face_on_image = face[0]
+ image, face, crop_face = futils.dlib.crop(image, face_on_image, self.up_ratio, self.down_ratio,
+ self.width_ratio)
+ np_image = np.array(image)
+ mask = self.face_parser.parse(np.float32(cv2.resize(np_image, self.input_size)))
+ mask = cv2.resize(mask.numpy(), (256, 256))
+ mask = mask.astype(np.uint8)
+ mask = mask2image(mask)
+
+ return mask
diff --git a/modules/image/Image_gan/style_transfer/face_parse/module.py b/modules/image/Image_gan/style_transfer/face_parse/module.py
new file mode 100644
index 0000000000000000000000000000000000000000..13bfdad5cd9ba46d91e11b25d8071b1f542a4bfb
--- /dev/null
+++ b/modules/image/Image_gan/style_transfer/face_parse/module.py
@@ -0,0 +1,131 @@
+# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import argparse
+import copy
+
+import paddle
+import paddlehub as hub
+from paddlehub.module.module import moduleinfo, runnable, serving
+import numpy as np
+import cv2
+from skimage.io import imread
+from skimage.transform import rescale, resize
+
+from .model import FaceParsePredictor
+from .util import base64_to_cv2
+
+
+@moduleinfo(
+ name="face_parse", type="CV/style_transfer", author="paddlepaddle", author_email="", summary="", version="1.0.0")
+class Face_parse:
+ def __init__(self):
+ self.pretrained_model = os.path.join(self.directory, "bisenet.pdparams")
+
+ self.network = FaceParsePredictor()
+
+ def style_transfer(self,
+ images=None,
+ paths=None,
+ output_dir='./transfer_result/',
+ use_gpu=False,
+ visualization=True):
+ '''
+
+
+ images (list[numpy.ndarray]): data of images, shape of each is [H, W, C], color space must be BGR(read by cv2).
+ paths (list[str]): paths to images
+ output_dir: the dir to save the results
+ use_gpu: if True, use gpu to perform the computation, otherwise cpu.
+ visualization: if True, save results in output_dir.
+ '''
+ results = []
+ paddle.disable_static()
+ place = 'gpu:0' if use_gpu else 'cpu'
+ place = paddle.set_device(place)
+ if images == None and paths == None:
+ print('No image provided. Please input an image or a image path.')
+ return
+
+ if images != None:
+ for image in images:
+ image = image[:, :, ::-1]
+ out = self.network.run(image)
+ results.append(out)
+
+ if paths != None:
+ for path in paths:
+ image = cv2.imread(path)[:, :, ::-1]
+ out = self.network.run(image)
+ results.append(out)
+
+ if visualization == True:
+ if not os.path.exists(output_dir):
+ os.makedirs(output_dir, exist_ok=True)
+ for i, out in enumerate(results):
+ if out is not None:
+ cv2.imwrite(os.path.join(output_dir, 'output_{}.png'.format(i)), out[:, :, ::-1])
+
+ return results
+
+ @runnable
+ def run_cmd(self, argvs: list):
+ """
+ Run as a command.
+ """
+ self.parser = argparse.ArgumentParser(
+ description="Run the {} module.".format(self.name),
+ prog='hub run {}'.format(self.name),
+ usage='%(prog)s',
+ add_help=True)
+
+ self.arg_input_group = self.parser.add_argument_group(title="Input options", description="Input data. Required")
+ self.arg_config_group = self.parser.add_argument_group(
+ title="Config options", description="Run configuration for controlling module behavior, not required.")
+ self.add_module_config_arg()
+ self.add_module_input_arg()
+ self.args = self.parser.parse_args(argvs)
+ results = self.style_transfer(
+ paths=[self.args.input_path],
+ output_dir=self.args.output_dir,
+ use_gpu=self.args.use_gpu,
+ visualization=self.args.visualization)
+ return results
+
+ @serving
+ def serving_method(self, images, **kwargs):
+ """
+ Run as a service.
+ """
+ images_decode = [base64_to_cv2(image) for image in images]
+ results = self.style_transfer(images=images_decode, **kwargs)
+ tolist = [result.tolist() for result in results]
+ return tolist
+
+ def add_module_config_arg(self):
+ """
+ Add the command config options.
+ """
+ self.arg_config_group.add_argument('--use_gpu', action='store_true', help="use GPU or not")
+
+ self.arg_config_group.add_argument(
+ '--output_dir', type=str, default='transfer_result', help='output directory for saving result.')
+ self.arg_config_group.add_argument('--visualization', type=bool, default=False, help='save results or not.')
+
+ def add_module_input_arg(self):
+ """
+ Add the command input options.
+ """
+ self.arg_input_group.add_argument('--input_path', type=str, help="path to input image.")
diff --git a/modules/image/Image_gan/style_transfer/face_parse/requirements.txt b/modules/image/Image_gan/style_transfer/face_parse/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..d9bfc85782a3ee323241fe7beb87a9f281c120fe
--- /dev/null
+++ b/modules/image/Image_gan/style_transfer/face_parse/requirements.txt
@@ -0,0 +1,2 @@
+ppgan
+dlib
diff --git a/modules/image/Image_gan/style_transfer/face_parse/util.py b/modules/image/Image_gan/style_transfer/face_parse/util.py
new file mode 100644
index 0000000000000000000000000000000000000000..b88ac3562b74cadc1d4d6459a56097ca4a938a0b
--- /dev/null
+++ b/modules/image/Image_gan/style_transfer/face_parse/util.py
@@ -0,0 +1,10 @@
+import base64
+import cv2
+import numpy as np
+
+
+def base64_to_cv2(b64str):
+ data = base64.b64decode(b64str.encode('utf8'))
+ data = np.fromstring(data, np.uint8)
+ data = cv2.imdecode(data, cv2.IMREAD_COLOR)
+ return data