update

Dockerfile

@@ -1,35 +0,0 @@
-FROM ubuntu:22.04
-
-# apt换源，安装pip
-RUN echo "==> 换成清华源，并更新..." && \
- sed -i s@/archive.ubuntu.com/@/mirrors.tuna.tsinghua.edu.cn/@g /etc/apt/sources.list && \
- sed -i s@/security.ubuntu.com/@/mirrors.tuna.tsinghua.edu.cn/@g /etc/apt/sources.list && \
- apt-get clean && \
- apt-get update
-
-# 安装python3.10
-RUN apt-get install -y python3 curl && \
- curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py && \
- python3 get-pip.py && \
- pip3 install -U pip && \
- pip3 config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
-
-# 安装ffmpeg等库
-RUN apt-get install libpython3.10-dev ffmpeg libgl1-mesa-glx libglib2.0-0 cmake -y && \
- pip3 install --no-cache-dir cmake
-
-WORKDIR /app
-
-COPY . .
-
-RUN pip3 install -r requirements.txt && \
- pip3 install -r requirements-app.txt
-
-RUN echo "==> Clean up..." && \
- rm -rf ~/.cache/pip
-
-# 指定工作目录
-
-EXPOSE 7860
-
-CMD [ "python3", "app.py", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -7,328 +7,4 @@ sdk: gradio
 sdk_version: 4.43.0
 app_file: app.py
 pinned: true
----
-
-
-<div align="center">
-<h1>HivisionIDPhoto</h1>
-
-[English](README_EN.md) / 中文 / [日本語](README_JP.md) / [한국어](README_KO.md)
-
-[![GitHub](https://img.shields.io/static/v1?label=GitHub&message=GitHub&color=black)](https://github.com/xiaolin199912/HivisionIDPhotos)
-[![SwanHub Demo](https://swanhub.co/git/repo/SwanHub%2FAuto-README/file/preview?ref=main&path=swanhub.svg)](https://swanhub.co/ZeYiLin/HivisionIDPhotos/demo)
-[![zhihu](https://img.shields.io/static/v1?label=知乎&message=知乎&color=blue)](https://zhuanlan.zhihu.com/p/638254028)
-[![Spaces](https://img.shields.io/badge/🤗-Open%20in%20Spaces-blue)](https://huggingface.co/spaces/TheEeeeLin/HivisionIDPhotos)
-<a href="https://docs.qq.com/doc/DUkpBdk90eWZFS2JW" target="_blank">
-<img alt="Static Badge" src="https://img.shields.io/badge/WeChat-微信-4cb55e"></a>
-
-<a href="https://trendshift.io/repositories/11622" target="_blank"><img src="https://trendshift.io/api/badge/repositories/11622" alt="Zeyi-Lin%2FHivisionIDPhotos | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
-
-<img src="assets/demoImage.png" width=900>
-
-</div>
-
-<br>
-
-> **相关项目**：
->
-> - [SwanLab](https://github.com/SwanHubX/SwanLab)：训练人像抠图模型全程用它来分析和监控，以及和实验室同学协作交流，大幅提升了训练效率。
-
-<br>
-
-# 🤩 项目更新
-
-- 在线体验： [![SwanHub Demo](https://img.shields.io/static/v1?label=Demo&message=SwanHub%20Demo&color=blue)](https://swanhub.co/ZeYiLin/HivisionIDPhotos/demo)、[![Spaces](https://img.shields.io/badge/🤗-Open%20in%20Spaces-blue)](https://huggingface.co/spaces/TheEeeeLin/HivisionIDPhotos)
-
-- 2024.9.5: 更新 [Restful API 文档](docs/api_CN.md)
-- 2024.9.2: 更新**调整照片 KB 大小**，[DockerHub](https://hub.docker.com/r/linzeyi/hivision_idphotos/tags)
-- 2023.12.1: 更新**API 部署（基于 fastapi）**
-- 2023.6.20: 更新**预设尺寸菜单**
-- 2023.6.19: 更新**排版照**
-
-# Overview
-
-> 🚀 谢谢你对我们的工作感兴趣。您可能还想查看我们在图像领域的其他成果，欢迎来信:[email protected].
-
-HivisionIDPhoto 旨在开发一种实用的证件照智能制作算法。
-
-它利用一套完善的模型工作流程，实现对多种用户拍照场景的识别、抠图与证件照生成。
-
-**HivisionIDPhoto 可以做到：**
-
-1. 轻量级抠图（仅需 **CPU** 即可快速推理）
-2. 根据不同尺寸规格生成不同的标准证件照、六寸排版照
-3. 美颜（waiting）
-4. 智能换正装（waiting）
-
-<div align="center">
-<img src="assets/gradio-image.jpeg" width=900>
-</div>
-
----
-
-如果 HivisionIDPhoto 对你有帮助，请 star 这个 repo 或推荐给你的朋友，解决证件照应急制作问题！
-
-<br>
-
-# 🔧 环境安装与依赖
-
-- Python >= 3.7（项目主要测试在 python 3.10）
-- onnxruntime
-- OpenCV
-- Option: Linux, Windows, MacOS
-
-**1. 克隆项目**
-
-```bash
-git clone https://github.com/Zeyi-Lin/HivisionIDPhotos.git
-cd HivisionIDPhotos
-```
-
-**2. （重要）安装依赖环境**
-
-> 建议 conda 创建一个 python3.10 虚拟环境后，执行以下命令
-
-```bash
-pip install -r requirements.txt
-pip install -r requirements-app.txt
-```
-
-**3. 下载权重文件**
-
-在我们的[Release](https://github.com/Zeyi-Lin/HivisionIDPhotos/releases/tag/pretrained-model)下载权重文件`hivision_modnet.onnx` (24.7MB)，存到项目的`hivision/creator/weights`目录下。
-
-<br>
-
-# 🚀 运行 Gradio Demo
-
-```bash
-python app.py
-```
-
-运行程序将生成一个本地 Web 页面，在页面中可完成证件照的操作与交互。
-
-<br>
-
-# 🚀 Python 推理
-
-## 1. 证件照制作
-
-输入 1 张照片，获得 1 张标准证件照和 1 张高清证件照的 4 通道透明 png
-
-```python
-python inference.py -i demo/images/test.jpg -o ./idphoto.png --height 413 --width 295
-```
-
-## 2. 增加底色
-
-输入 1 张 4 通道透明 png，获得 1 张增加了底色的图像）
-
-```python
-python inference.py -t add_background -i ./idphoto.png -o ./idhoto_ab.jpg -c 000000 -k 30
-```
-
-## 3. 得到六寸排版照
-
-输入 1 张 3 通道照片，获得 1 张六寸排版照
-
-```python
-python inference.py -t generate_layout_photos -i ./idhoto_ab.jpg -o ./idhoto_layout.jpg --height 413 --width 295 -k 200
-```
-
-<br>
-
-# ⚡️ 部署 API 服务
-
-## 启动后端
-
-```
-python deploy_api.py
-```
-
-
-## 请求 API 服务 - Python Request
-
-> 请求方式请参考 [API 文档](docs/api_CN.md)，含 [cURL](docs/api_CN.md#curl-请求示例)、[Python](docs/api_CN.md#python-请求示例)、[Java](docs/api_CN.md#java-请求示例)、[Javascript](docs/api_CN.md#javascript-请求示例) 请求示例。
-
-### 1. 证件照制作
-
-输入 1 张照片，获得 1 张标准证件照和 1 张高清证件照的 4 通道透明 png
-
-```python
-import requests
-
-url = "http://127.0.0.1:8080/idphoto"
-input_image_path = "demo/images/test.jpg"
-
-files = {"input_image": open(input_image_path, "rb")}
-data = {"height": 413, "width": 295}
-
-response = requests.post(url, files=files, data=data).json()
-
-# response为一个json格式字典，包含status、image_base64_standard和image_base64_hd三项
-print(response)
-
-```
-
-### 2. 增加底色
-
-输入 1 张 4 通道透明 png，获得 1 张增加了底色的图像
-
-```python
-import requests
-
-url = "http://127.0.0.1:8080/add_background"
-input_image_path = "test.png"
-
-files = {"input_image": open(input_image_path, "rb")}
-data = {"color": '638cce', 'kb': None}
-
-response = requests.post(url, files=files, data=data).json()
-
-# response为一个json格式字典，包含status和image_base64
-print(response)
-```
-
-### 3. 得到六寸排版照
-
-输入 1 张 3 通道照片，获得 1 张六寸排版照
-
-```python
-import requests
-
-url = "http://127.0.0.1:8080/generate_layout_photos"
-input_image_path = "test.jpg"
-
-files = {"input_image": open(input_image_path, "rb")}
-data = {"height": 413, "width": 295, "kb": 200}
-
-response = requests.post(url, files=files, data=data).json()
-
-# response为一个json格式字典，包含status和image_base64
-print(response)
-```
-
-<br>
-
-# 🐳 Docker 部署
-
-## 1. 拉取或构建镜像
-
-> 以下方式三选一
-
-**方式一：拉取镜像：**
-
-```bash
-docker pull linzeyi/hivision_idphotos:v1
-docker tag linzeyi/hivision_idphotos:v1 hivision_idphotos
-```
-
-国内拉取加速：
-
-```bash
-docker pull registry.cn-hangzhou.aliyuncs.com/swanhub/hivision_idphotos:v1
-docker tag registry.cn-hangzhou.aliyuncs.com/swanhub/hivision_idphotos:v1 hivision_idphotos
-```
-
-**方式二：Dockrfile 直接构建镜像：**
-
-在确保将模型权重文件[hivision_modnet.onnx](https://github.com/Zeyi-Lin/HivisionIDPhotos/releases/tag/pretrained-model)放到`hivision/creator/weights`下后，在项目根目录执行：
-
-```bash
-docker build -t hivision_idphotos .
-```
-
-**方式三：Docker compose 构建：**
-
-确保将模型权重文件 [hivision_modnet.onnx](https://github.com/Zeyi-Lin/HivisionIDPhotos/releases/tag/pretrained-model) 放在`hivision/creator/weights`下后，在项目根目录下执行：
-
-```bash
-docker compose build
-```
-
-镜像打包完成后，运行以下命令启动 Gradio 服务：
-
-```bash
-docker compose up -d
-```
-
-## 2. 运行 Gradio Demo
-
-等待镜像封装完毕后，运行以下指令，即可开启 Gradio Demo 服务：
-
-```bash
-docker run -p 7860:7860 hivision_idphotos
-```
-
-在你的本地访问 [http://127.0.0.1:7860](http://127.0.0.1:7860/) 即可使用。
-
-## 3. 运行 API 后端服务
-
-```bash
-docker run -p 8080:8080 hivision_idphotos python3 deploy_api.py
-```
-
-<br>
-
-# 🌲 友情链接
-
-- [HivisionIDPhotos-windows-GUI](https://github.com/zhaoyun0071/HivisionIDPhotos-windows-GUI)
-
-<br>
-
-# 📖 引用项目
-
-1. MTCNN:
-
-```bibtex
-@software{ipazc_mtcnn_2021,
- author = {ipazc},
- title = {{MTCNN}},
- url = {https://github.com/ipazc/mtcnn},
- year = {2021},
- publisher = {GitHub}
-}
-```
-
-2. ModNet:
-
-```bibtex
-@software{zhkkke_modnet_2021,
- author = {ZHKKKe},
- title = {{ModNet}},
- url = {https://github.com/ZHKKKe/MODNet},
- year = {2021},
- publisher = {GitHub}
-}
-```
-
-<br>
-
-# 💻 开发小贴士
-
-**1. 如何修改预设尺寸？**
-
-修改[size_list_CN.csv](demo/size_list_CN.csv)后再次运行 app.py 即可，其中第一列为尺寸名，第二列为高度，第三列为宽度。
-
-<br>
-
-# 📧 联系我们
-
-如果您有任何问题，请发邮件至 [email protected]
-
-<br>
-
-# 贡献者
-
-<a href="https://github.com/Zeyi-Lin/HivisionIDPhotos/graphs/contributors">
- <img src="https://contrib.rocks/image?repo=Zeyi-Lin/HivisionIDPhotos" />
-</a>
-
-[Zeyi-Lin](https://github.com/Zeyi-Lin)、[SAKURA-CAT](https://github.com/SAKURA-CAT)、[Feudalman](https://github.com/Feudalman)、[swpfY](https://github.com/swpfY)、[Kaikaikaifang](https://github.com/Kaikaikaifang)、[ShaohonChen](https://github.com/ShaohonChen)、[KashiwaByte](https://github.com/KashiwaByte)
-
-<br>
-
-# StarHistory
-
-[![Star History Chart](https://api.star-history.com/svg?repos=Zeyi-Lin/HivisionIDPhotos&type=Date)](https://star-history.com/#Zeyi-Lin/HivisionIDPhotos&Date)
+---

app.py

@@ -1,16 +1,14 @@
 import os
 import gradio as gr
 from hivision import IDCreator
-from hivision.error import FaceError
+from hivision.error import FaceError, APIError
 from hivision.utils import add_background, resize_image_to_kb
 from hivision.creator.layout_calculator import (
  generate_layout_photo,
  generate_layout_image,
 )
-from hivision.creator.human_matting import (
- extract_human_modnet_photographic_portrait_matting,
- extract_human,
-)
+from hivision.creator.human_matting import *
+from hivision.creator.face_detector import *
 import pathlib
 import numpy as np
 from demo.utils import csv_to_size_list
@@ -59,6 +57,7 @@ def idphoto_inference(
  custom_image_kb,
  language,
  matting_model_option,
+ face_detect_option,
  head_measure_ratio=0.2,
  head_height_ratio=0.45,
  top_distance_max=0.12,
@@ -78,7 +77,7 @@ def idphoto_inference(
  "The width should not be greater than the length; the length and width should not be less than 100, and no more than 1800.": "宽度应不大于长度；长宽不应小于 100，大于 1800",
  "Custom Color": "自定义底色",
  "Custom": "自定义",
- "The number of faces is not equal to 1": "人脸数量不等于 1",
+ "The number of faces is not equal to 1, please upload an image with a single face. If the actual number of faces is 1, it may be an issue with the accuracy of the detection model. Please switch to a different face detection model on the left or raise a Github Issue to notify the author.": "人脸数量不等于 1，请上传单张人脸的图像。如果实际人脸数量为 1，可能是检测模型精度的问题，请在左边更换人脸检测模型或给作者提Github Issue。",
  "Solid Color": "纯色",
  "Up-Down Gradient (White)": "上下渐变 (白)",
  "Center Gradient (White)": "中心渐变 (白)",
@@ -92,7 +91,7 @@ def idphoto_inference(
  "The width should not be greater than the length; the length and width should not be less than 100, and no more than 1800.": "The width should not be greater than the length; the length and width should not be less than 100, and no more than 1800.",
  "Custom Color": "Custom Color",
  "Custom": "Custom",
- "The number of faces is not equal to 1": "The number of faces is not equal to 1",
+ "The number of faces is not equal to 1, please upload an image with a single face. If the actual number of faces is 1, it may be an issue with the accuracy of the detection model. Please switch to a different face detection model on the left or raise a Github Issue to notify the author.": "The number of faces is not equal to 1, please upload an image with a single face. If the actual number of faces is 1, it may be an issue with the accuracy of the detection model. Please switch to a different face detection model on the left or raise a Github Issue to notify the author.",
  "Solid Color": "Solid Color",
  "Up-Down Gradient (White)": "Up-Down Gradient (White)",
  "Center Gradient (White)": "Center Gradient (White)",
@@ -156,6 +155,11 @@ def idphoto_inference(
  else:
  creator.matting_handler = extract_human
 
+ if face_detect_option == "mtcnn":
+ creator.detection_handler = detect_face_mtcnn
+ else:
+ creator.detection_handler = detect_face_face_plusplus
+
  change_bg_only = idphoto_json["size_mode"] in ["只换底", "Only Change Background"]
  # 生成证件照
  try:
@@ -170,8 +174,21 @@ def idphoto_inference(
  result_message = {
  img_output_standard: gr.update(value=None),
  img_output_standard_hd: gr.update(value=None),
+ img_output_layout: gr.update(visible=False),
+ notification: gr.update(
+ value=text_lang_map[language][
+ "The number of faces is not equal to 1, please upload an image with a single face. If the actual number of faces is 1, it may be an issue with the accuracy of the detection model. Please switch to a different face detection model on the left or raise a Github Issue to notify the author."
+ ],
+ visible=True,
+ ),
+ }
+ except APIError as e:
+ result_message = {
+ img_output_standard: gr.update(value=None),
+ img_output_standard_hd: gr.update(value=None),
+ img_output_layout: gr.update(visible=False),
  notification: gr.update(
- value=text_lang_map[language]["The number of faces is not equal to 1"],
+ value=f"Please make sure you have correctly set up the Face++ API Key and Secret.\nAPI Error\nStatus Code is {e.status_code}\nPossible errors are: {e}\n",
  visible=True,
  ),
  }
@@ -284,6 +301,12 @@ if __name__ == "__main__":
  # argparser.add_argument(
  # "--host", type=str, default="127.0.0.1", help="The host of the server"
  # )
+ # argparser.add_argument(
+ # "--root_path",
+ # type=str,
+ # default=None,
+ # help="The root path of the server, default is None (='/'), e.g. '/myapp'",
+ # )
 
  # args = argparser.parse_args()
 
@@ -294,6 +317,12 @@ if __name__ == "__main__":
  for file in os.listdir(os.path.join(root_dir, "hivision/creator/weights"))
  if file.endswith(".onnx")
  ]
+ DEFAULT_MATTING_MODEL = "modnet_photographic_portrait_matting"
+ if DEFAULT_MATTING_MODEL in matting_model_list:
+ matting_model_list.remove(DEFAULT_MATTING_MODEL)
+ matting_model_list.insert(0, DEFAULT_MATTING_MODEL)
+
+ face_detect_model_list = ["mtcnn", "face++ (联网API)"]
 
  size_mode_CN = ["尺寸列表", "只换底", "自定义尺寸"]
  size_mode_EN = ["Size List", "Only Change Background", "Custom Size"]
@@ -353,10 +382,16 @@ if __name__ == "__main__":
  value="中文",
  elem_id="language",
  )
+ face_detect_model_options = gr.Dropdown(
+ choices=face_detect_model_list,
+ label="人脸检测模型",
+ value=face_detect_model_list[0],
+ elem_id="matting_model",
+ )
  matting_model_options = gr.Dropdown(
  choices=matting_model_list,
- label="Matting Model",
- value="modnet_photographic_portrait_matting",
+ label="抠图模型",
+ value=matting_model_list[0],
  elem_id="matting_model",
  )
 
@@ -482,6 +517,8 @@ if __name__ == "__main__":
  choices=image_kb_CN,
  value="不设置",
  ),
+ matting_model_options: gr.update(label="抠图模型"),
+ face_detect_model_options: gr.update(label="人脸检测模型"),
  custom_image_kb_size: gr.update(label="KB 大小"),
  notification: gr.update(label="状态"),
  img_output_standard: gr.update(label="标准照"),
@@ -518,6 +555,8 @@ if __name__ == "__main__":
  choices=image_kb_EN,
  value="Not Set",
  ),
+ matting_model_options: gr.update(label="Matting model"),
+ face_detect_model_options: gr.update(label="Face detect model"),
  custom_image_kb_size: gr.update(label="KB size"),
  notification: gr.update(label="Status"),
  img_output_standard: gr.update(label="Standard photo"),
@@ -574,6 +613,8 @@ if __name__ == "__main__":
  img_but,
  render_options,
  image_kb_options,
+ matting_model_options,
+ face_detect_model_options,
  custom_image_kb_size,
  notification,
  img_output_standard,
@@ -614,6 +655,7 @@ if __name__ == "__main__":
  custom_image_kb_size,
  language_options,
  matting_model_options,
+ face_detect_model_options,
  ],
  outputs=[
  img_output_standard,
@@ -624,4 +666,10 @@ if __name__ == "__main__":
  ],
  )
 
- demo.launch()
+ demo.launch(
+ # server_name=args.host,
+ # server_port=args.port,
+ # show_api=False,
+ # favicon_path=os.path.join(root_dir, "assets/hivision_logo.png"),
+ # root_path=args.root_path,
+ )

app.spec

@@ -1,44 +0,0 @@
-# -*- mode: python ; coding: utf-8 -*-
-from PyInstaller.utils.hooks import collect_data_files
-
-datas = [('hivisionai', 'hivisionai'), ('hivision_modnet.onnx', '.'), ('size_list_CN.csv', '.')]
-datas += collect_data_files('gradio_client')
-datas += collect_data_files('gradio')
-
-
-a = Analysis(
- ['app/web.py'],
- pathex=[],
- binaries=[],
- datas=datas,
- hiddenimports=[],
- hookspath=[],
- hooksconfig={},
- runtime_hooks=[],
- excludes=[],
- noarchive=False,
- optimize=0,
-)
-pyz = PYZ(a.pure)
-
-exe = EXE(
- pyz,
- a.scripts,
- a.binaries,
- a.datas,
- [],
- name='HivisionIDPhotos',
- debug=False,
- bootloader_ignore_signals=False,
- strip=False,
- upx=True,
- upx_exclude=[],
- runtime_tmpdir=None,
- console=True,
- disable_windowed_traceback=False,
- argv_emulation=False,
- target_arch=None,
- codesign_identity=None,
- entitlements_file=None,
- icon=['assets\hivisionai.ico'],
-)

deploy_api.py

@@ -1,157 +0,0 @@
-from fastapi import FastAPI, UploadFile, Form
-import onnxruntime
-from hivision import IDCreator
-from hivision.error import FaceError
-from hivision.creator.layout_calculator import (
- generate_layout_photo,
- generate_layout_image,
-)
-from hivision.utils import add_background, resize_image_to_kb_base64, hex_to_rgb
-import base64
-import numpy as np
-import cv2
-import os
-
-app = FastAPI()
-creator = IDCreator()
-
-
-# 将图像转换为Base64编码
-def numpy_2_base64(img: np.ndarray):
- retval, buffer = cv2.imencode(".png", img)
- base64_image = base64.b64encode(buffer).decode("utf-8")
-
- return base64_image
-
-
-# 证件照智能制作接口
-@app.post("/idphoto")
-async def idphoto_inference(
- input_image: UploadFile,
- height: str = Form(...),
- width: str = Form(...),
- head_measure_ratio=0.2,
- head_height_ratio=0.45,
- top_distance_max=0.12,
- top_distance_min=0.10,
-):
-
- image_bytes = await input_image.read()
- nparr = np.frombuffer(image_bytes, np.uint8)
- img = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
-
- # 将字符串转为元组
- size = (int(height), int(width))
- try:
- result = creator(
- img,
- size=size,
- head_measure_ratio=head_measure_ratio,
- head_height_ratio=head_height_ratio,
- )
- except FaceError:
- result_message = {"status": False}
- # 如果检测到人脸数量等于1, 则返回标准证和高清照结果（png 4通道图像）
- else:
- result_message = {
- "status": True,
- "image_base64_standard": numpy_2_base64(result.standard),
- "image_base64_hd": numpy_2_base64(result.hd),
- }
-
- return result_message
-
-
-# 透明图像添加纯色背景接口
-@app.post("/add_background")
-async def photo_add_background(
- input_image: UploadFile, color: str = Form(...), kb: str = Form(None)
-):
- image_bytes = await input_image.read()
- nparr = np.frombuffer(image_bytes, np.uint8)
- img = cv2.imdecode(nparr, cv2.IMREAD_UNCHANGED)
-
- color = hex_to_rgb(color)
- color = (color[2], color[1], color[0])
-
- result_image = add_background(img, bgr=color).astype(np.uint8)
-
- if kb:
- result_image = cv2.cvtColor(result_image, cv2.COLOR_RGB2BGR)
- result_image_base64 = resize_image_to_kb_base64(result_image, int(kb))
- else:
- result_image_base64 = numpy_2_base64(result_image)
-
- # try:
-
- result_messgae = {
- "status": True,
- "image_base64": result_image_base64,
- }
-
- # except Exception as e:
- # print(e)
- # result_messgae = {
- # "status": False,
- # "error": e
- # }
-
- return result_messgae
-
-
-# 六寸排版照生成接口
-@app.post("/generate_layout_photos")
-async def generate_layout_photos(
- input_image: UploadFile,
- height: str = Form(...),
- width: str = Form(...),
- kb: str = Form(None),
-):
- # try:
- image_bytes = await input_image.read()
- nparr = np.frombuffer(image_bytes, np.uint8)
- img = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
-
- size = (int(height), int(width))
-
- typography_arr, typography_rotate = generate_layout_photo(
- input_height=size[0], input_width=size[1]
- )
-
- result_layout_image = generate_layout_image(
- img, typography_arr, typography_rotate, height=size[0], width=size[1]
- ).astype(np.uint8)
-
- if kb:
- result_layout_image = cv2.cvtColor(result_layout_image, cv2.COLOR_RGB2BGR)
- result_layout_image_base64 = resize_image_to_kb_base64(
- result_layout_image, int(kb)
- )
- else:
- result_layout_image_base64 = numpy_2_base64(result_layout_image)
-
- result_messgae = {
- "status": True,
- "image_base64": result_layout_image_base64,
- }
-
- # except Exception as e:
- # result_messgae = {
- # "status": False,
- # }
-
- return result_messgae
-
-
-if __name__ == "__main__":
- import uvicorn
-
- # 加载权重文件
- root_dir = os.path.dirname(os.path.abspath(__file__))
- HY_HUMAN_MATTING_WEIGHTS_PATH = os.path.join(
- root_dir, "hivision/creator/weights/hivision_modnet.onnx"
- )
- sess = onnxruntime.InferenceSession(HY_HUMAN_MATTING_WEIGHTS_PATH)
-
- # 在8080端口运行推理服务
- uvicorn.run(app, host="0.0.0.0", port=8080)

docker-compose.yml

@@ -1,11 +0,0 @@
-version: "3.8"
-
-services:
- hivision_idphotos:
- build:
- context: .
- dockerfile: Dockerfile
- image: hivision_idphotos2
- command: python3 app.py --host 0.0.0.0 --port 7860
- ports:
- - "7860:7860"

docs/api_CN.md

@@ -1,554 +0,0 @@
-# API Docs
-
-中文 | [English](api_EN.md)
-
-## 目录
-
-- [开始之前：开启后端服务](#开始之前开启后端服务)
-- [接口功能说明](#接口功能说明)
-- [cURL 请求示例](#curl-请求示例)
-- [Python 请求示例](#python-请求示例)
- - [Python Requests 请求方法](#1️⃣-python-requests-请求方法)
- - [Python 脚本请求方法](#2️⃣-python-脚本请求方法)
-- [Java 请求示例](#java-请求示例)
-- [Javascript 请求示例](#javascript-请求示例)
-
-## 开始之前：开启后端服务
-
-在请求 API 之前，请先运行后端服务
-
-```bash
-python delopy_api.py
-```
-
-<br>
-
-## 接口功能说明
-
-### 1.生成证件照(底透明)
-
-接口名：`idphoto`
-
-`生成证件照`接口的逻辑是发送一张 RGB 图像，输出一张标准证件照和一张高清证件照：
-
-- **高清证件照**：根据`size`的宽高比例制作的证件照，文件名为`output_image_dir`增加`_hd`后缀
-- **标准证件照**：尺寸等于`size`，由高清证件照缩放而来，文件名为`output_image_dir`
-
-需要注意的是，生成的两张照片都是透明的（RGBA 四通道图像），要生成完整的证件照，还需要下面的`添加背景色`接口。
-
-> 问：为什么这么设计？ 
-> 答：因为在实际产品中，经常用户会频繁切换底色预览效果，直接给透明底图像，由前端 js 代码合成颜色是更好体验的做法。
-
-### 2.添加背景色
-
-接口名：`add_background`
-
-`添加背景色`接口的逻辑是发送一张 RGBA 图像，根据`color`添加背景色，合成一张 JPG 图像。
-
-### 3.生成六寸排版照
-
-接口名：`generate_layout_photos`
-
-`生成六寸排版照`接口的逻辑是发送一张 RGB 图像（一般为添加背景色之后的证件照），根据`size`进行照片排布，然后生成一张六寸排版照。
-
-<br>
-
-
-## cURL 请求示例
-
-cURL 是一个命令行工具，用于使用各种网络协议传输数据。以下是使用 cURL 调用这些 API 的示例。
-
-### 1. 生成证件照(底透明)
-
-```bash
-curl -X POST "http://127.0.0.1:8080/idphoto" \
--F "input_image=@demo/images/test.jpg" \
--F "height=413" \
--F "width=295"
-```
-
-### 2. 添加背景色
-
-```bash
-curl -X POST "http://127.0.0.1:8080/add_background" \
--F "[email protected]" \
--F "color=638cce" \
--F "kb=200"
-```
-
-### 3. 生成六寸排版照
-
-```bash
-curl -X POST "http://127.0.0.1:8080/generate_layout_photos" \
--F "[email protected]" \
--F "height=413" \
--F "width=295" \
--F "kb=200"
-```
-
-
-## Python 请求示例
-
-### 1️⃣ Python Requests 请求方法
-
-#### 1.生成证件照(底透明)
-
-```python
-import requests
-
-url = "http://127.0.0.1:8080/idphoto"
-input_image_path = "images/test.jpg"
-
-files = {"input_image": open(input_image_path, "rb")}
-data = {"height": 413, "width": 295}
-
-response = requests.post(url, files=files, data=data).json()
-
-# response为一个json格式字典，包含status、image_base64_standard和image_base64_hd三项
-print(response)
-
-```
-
-#### 2.添加背景色
-
-```python
-import requests
-
-url = "http://127.0.0.1:8080/add_background"
-input_image_path = "test.png"
-
-files = {"input_image": open(input_image_path, "rb")}
-data = {"color": '638cce', 'kb': None}
-
-response = requests.post(url, files=files, data=data).json()
-
-# response为一个json格式字典，包含status和image_base64
-print(response)
-```
-
-#### 3.生成六寸排版照
-
-```python
-import requests
-
-url = "http://127.0.0.1:8080/generate_layout_photos"
-input_image_path = "test.jpg"
-
-files = {"input_image": open(input_image_path, "rb")}
-data = {"height": 413, "width": 295, "kb": 200}
-
-response = requests.post(url, files=files, data=data).json()
-
-# response为一个json格式字典，包含status和image_base64
-print(response)
-```
-
-<br>
-
-### 2️⃣ Python 脚本请求方法
-
-```bash
-python requests_api.py -u <URL> -t <TYPE> -i <INPUT_IMAGE_DIR> -o <OUTPUT_IMAGE_DIR> [--height <HEIGHT>] [--width <WIDTH>] [-c <COLOR>] [-k <KB>]
-```
-
-#### 参数说明
-
-##### 基本参数
-
-- `-u`, `--url`
-
- - **描述**: API 服务的 URL。
- - **默认值**: `http://127.0.0.1:8080`
-
-- `-t`, `--type`
-
- - **描述**: 请求 API 的种类，可选值有 `idphoto`、`add_background` 和 `generate_layout_photos`。分别代表证件照制作、透明图加背景和排版照生成。
- - **默认值**: `idphoto`
-
-- `-i`, `--input_image_dir`
-
- - **描述**: 输入图像路径。
- - **必需**: 是
- - **示例**: `./input_images/photo.jpg`
-
-- `-o`, `--output_image_dir`
- - **描述**: 保存图像路径。
- - **必需**: 是
- - **示例**: `./output_images/processed_photo.jpg`
-
-##### 可选参数
-
-- `--height`,
- - **描述**: 标准证件照的输出尺寸的高度。
- - **默认值**: 413
-- `--width`,
-
- - **描述**: 标准证件照的输出尺寸的宽度。
- - **默认值**: 295
-
-- `-c`, `--color`
-
- - **描述**: 给透明图增加背景色，格式为 Hex（如#638cce），仅在 type 为`add_background`时生效
- - **默认值**: `638cce`
-
-- `-k`, `--kb`
- - **描述**: 输出照片的 KB 值，仅在 type 为`add_background`和`generate_layout_photos`时生效，值为 None 时不做设置。
- - **默认值**: `None`
- - **示例**: `50`
-
-### 1.生成证件照(底透明)
-
-```bash
-python requests_api.py \
- -u http://127.0.0.1:8080 \
- -t idphoto \
- -i ./photo.jpg \
- -o ./idphoto.png \
- --height 413 \
- --width 295
-```
-
-### 2.添加背景色
-
-```bash
-python requests_api.py \
- -u http://127.0.0.1:8080 \
- -t add_background \
- -i ./idphoto.png \
- -o ./idphoto_with_background.jpg \
- -c 638cce \
- -k 50
-```
-
-### 3.生成六寸排版照
-
-```bash
-python requests_api.py \
- -u http://127.0.0.1:8080 \
- -t generate_layout_photos \
- -i ./idphoto_with_background.jpg \
- -o ./layout_photo.jpg \
- --height 413 \
- --width 295 \
- -k 200
-```
-
-### 请求失败的情况
-
-- 照片中检测到的人脸大于 1，则失败
-
-## Java 请求示例
-
-### 添加 maven 依赖
-
-```java
- <dependency>
- <groupId>cn.hutool</groupId>
- <artifactId>hutool-all</artifactId>
- <version>5.8.16</version>
- </dependency>
-
- <dependency>
- <groupId>commons-io</groupId>
- <artifactId>commons-io</artifactId>
- <version>2.6</version>
- </dependency>
-```
-
-### 运行代码
-
-#### 1.生成证件照(底透明)
-
-```java
- /**
- * 生成证件照(底透明) /idphoto 接口
- * @param inputImageDir 文件地址
- * @return
- * @throws IOException
- */
- public static String requestIdPhoto(String inputImageDir) throws IOException {
- String url = BASE_URL+"/idphoto";
- // 创建文件对象
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("height","413");
- paramMap.put("width","295");
- //包含status、image_base64_standard和image_base64_hd三项
- return HttpUtil.post(url, paramMap);
- }
-```
-
-#### 2.添加背景色
-
-```java
- /**
- * 添加背景色 /add_background 接口
- * @param inputImageDir 文件地址
- * @return
- * @throws IOException
- */
- public static String requestAddBackground(String inputImageDir) throws IOException {
- String url = BASE_URL+"/add_background";
- // 创建文件对象
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("color","638cce");
- paramMap.put("kb","200");
- // response为一个json格式字典，包含status和image_base64
- return HttpUtil.post(url, paramMap);
- }
-```
-
-#### 3.生成六寸排版照
-
-```java
- /**
- * 生成六寸排版照 /generate_layout_photos 接口
- * @param inputImageDir 文件地址
- * @return
- * @throws IOException
- */
- public static String requestGenerateLayoutPhotos(String inputImageDir) throws IOException {
- String url = BASE_URL+"/generate_layout_photos";
- // 创建文件对象
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("height","413");
- paramMap.put("width","295");
- paramMap.put("kb","200");
- //response为一个json格式字典，包含status和image_base64
- return HttpUtil.post(url, paramMap);
- }
-```
-
-#### 4.汇总
-
-```java
-
-import cn.hutool.http.HttpUtil;
-import cn.hutool.json.JSONObject;
-import cn.hutool.json.JSONUtil;
-import org.apache.commons.io.FileUtils;
-import org.springframework.util.StringUtils;
-import java.io.File;
-import java.io.IOException;
-import java.util.Base64;
-import java.util.HashMap;
-import java.util.Map;
-
-/**
- * @author: qingshuang
- * @createDate: 2024/09/05
- * @description: java生成证件照，测试用例
- */
-public class Test {
- /**
- * 接口地址
- */
- private final static String BASE_URL = "http://127.0.0.1:8080";
-
- /**
- * 生成证件照(底透明) /idphoto 接口
- * @param inputImageDir 文件地址
- * @return
- * @throws IOException
- */
- public static String requestIdPhoto(String inputImageDir) throws IOException {
- String url = BASE_URL+"/idphoto";
- // 创建文件对象
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("height","413");
- paramMap.put("width","295");
- return HttpUtil.post(url, paramMap);
- }
- /**
- * 添加背景色 /add_background 接口
- * @param inputImageDir 文件地址
- * @return
- * @throws IOException
- */
- public static String requestAddBackground(String inputImageDir) throws IOException {
- String url = BASE_URL+"/add_background";
- // 创建文件对象
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("color","638cce");
- paramMap.put("kb","200");
- return HttpUtil.post(url, paramMap);
- }
- /**
- * 生成六寸排版照 /generate_layout_photos 接口
- * @param inputImageDir 文件地址
- * @return
- * @throws IOException
- */
- public static String requestGenerateLayoutPhotos(String inputImageDir) throws IOException {
- String url = BASE_URL+"/generate_layout_photos";
- // 创建文件对象
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("height","413");
- paramMap.put("width","295");
- paramMap.put("kb","200");
- return HttpUtil.post(url, paramMap);
- }
- /**
- * 生成证件照(底透明)
- * @param inputImageDir 源文件地址
- * @param outputImageDir 输出文件地址
- * @throws IOException
- */
- private static void requestIdPhotoToImage(String inputImageDir, String outputImageDir) throws IOException {
- String res =requestIdPhoto(inputImageDir);
- //转成json
- JSONObject response= JSONUtil.parseObj(res);
- if(response.getBool("status")){//请求接口成功
- String image_base64_standard= response.getStr("image_base64_standard");
- String image_base64_hd =response.getStr("image_base64_hd");
- String[] outputImageDirArr= StringUtils.split(outputImageDir,".");
- // Base64 保存为图片
- FileUtils.writeByteArrayToFile(new File(outputImageDirArr[0]+"_standard."+outputImageDirArr[1]), Base64.getDecoder().decode(image_base64_standard));
- FileUtils.writeByteArrayToFile(new File(outputImageDirArr[0]+"_hd."+outputImageDirArr[1]), Base64.getDecoder().decode(image_base64_hd));
- }
- }
- /**
- * 添加背景色
- * @param inputImageDir 源文件地址
- * @param outputImageDir 输出文件地址
- * @throws IOException
- */
- private static void requestAddBackgroundToImage(String inputImageDir, String outputImageDir) throws IOException {
- String res =requestAddBackground(inputImageDir);
- //转成json
- JSONObject response= JSONUtil.parseObj(res);
- if(response.getBool("status")){//请求接口成功
- String image_base64= response.getStr("image_base64");
- String[] outputImageDirArr= StringUtils.split(outputImageDir,".");
- // Base64 保存为图片
- FileUtils.writeByteArrayToFile(new File(outputImageDirArr[0]+"_background."+outputImageDirArr[1]), Base64.getDecoder().decode(image_base64));
- }
- }
- /**
- * 生成六寸排版照
- * @param inputImageDir 源文件地址
- * @param outputImageDir 输出文件地址
- * @throws IOException
- */
- private static void requestGenerateLayoutPhotosToImage(String inputImageDir, String outputImageDir) throws IOException {
- String res =requestGenerateLayoutPhotos(inputImageDir);
- //转成json
- JSONObject response= JSONUtil.parseObj(res);
- if(response.getBool("status")){//请求接口成功
- String image_base64= response.getStr("image_base64");
- String[] outputImageDirArr= StringUtils.split(outputImageDir,".");
- // Base64 保存为图片
- FileUtils.writeByteArrayToFile(new File(outputImageDirArr[0]+"_layout."+outputImageDirArr[1]), Base64.getDecoder().decode(image_base64));
- }
- }
-
- public static void main(String[] args) {
- try {
- //生成证件照(底透明)
- requestIdPhotoToImage("C:\\Users\\Administrator\\Desktop\\1111.jpg","C:\\Users\\Administrator\\Desktop\\2222.png");
- //添加背景色
- requestAddBackgroundToImage("C:\\Users\\Administrator\\Desktop\\2222_hd.png","C:\\Users\\Administrator\\Desktop\\idphoto_with_background.jpg");
- //生成六寸排版照
- requestGenerateLayoutPhotosToImage("C:\\Users\\Administrator\\Desktop\\1111.jpg","C:\\Users\\Administrator\\Desktop\\2222.png");
-
- } catch (IOException e) {
- e.printStackTrace();
- }
- }
-}
-
-```
-
-## JavaScript 请求示例
-
-在JavaScript中，我们可以使用`fetch` API来发送HTTP请求。以下是如何使用JavaScript调用这些API的示例。
-
-### 1. 生成证件照(底透明)
-
-```javascript
-async function generateIdPhoto(inputImagePath, height, width) {
- const url = "http://127.0.0.1:8080/idphoto";
- const formData = new FormData();
- formData.append("input_image", new File([await fetch(inputImagePath).then(res => res.blob())], "test.jpg"));
- formData.append("height", height);
- formData.append("width", width);
-
- const response = await fetch(url, {
- method: 'POST',
- body: formData
- });
-
- const result = await response.json();
- console.log(result);
- return result;
-}
-
-// 示例调用
-generateIdPhoto("images/test.jpg", 413, 295).then(response => {
- console.log(response);
-});
-```
-
-### 2. 添加背景色
-
-```javascript
-async function addBackground(inputImagePath, color, kb) {
- const url = "http://127.0.0.1:8080/add_background";
- const formData = new FormData();
- formData.append("input_image", new File([await fetch(inputImagePath).then(res => res.blob())], "test.png"));
- formData.append("color", color);
- formData.append("kb", kb);
-
- const response = await fetch(url, {
- method: 'POST',
- body: formData
- });
-
- const result = await response.json();
- console.log(result);
- return result;
-}
-
-// 示例调用
-addBackground("test.png", "638cce", 200).then(response => {
- console.log(response);
-});
-```
-
-### 3. 生成六寸排版照
-
-```javascript
-async function generateLayoutPhotos(inputImagePath, height, width, kb) {
- const url = "http://127.0.0.1:8080/generate_layout_photos";
- const formData = new FormData();
- formData.append("input_image", new File([await fetch(inputImagePath).then(res => res.blob())], "test.jpg"));
- formData.append("height", height);
- formData.append("width", width);
- formData.append("kb", kb);
-
- const response = await fetch(url, {
- method: 'POST',
- body: formData
- });
-
- const result = await response.json();
- console.log(result);
- return result;
-}
-
-// 示例调用
-generateLayoutPhotos("test.jpg", 413, 295, 200).then(response => {
- console.log(response);
-});
-```

docs/api_EN.md

@@ -1,553 +0,0 @@
-# API Documentation
-
-[中文](api_CN.md) | English
-
-## TOC
-
-- [Before You Start: Launch the Backend Service](#before-you-start-launch-the-backend-service)
-- [Interface Function Descriptions](#interface-function-descriptions)
-- [cURL Request Example](#curl-request-examples)
-- [Python Request Example](#python-request-example)
- - [Python Requests Method](#1️⃣-python-requests-method)
- - [Python Script Method](#2️⃣-python-script-request-method)
-- [Java Request Example](#java-request-example)
-- [Javascript Request Example](#javascript-request-examples)
-
-## Before You Start: Launch the Backend Service
-
-Before making API requests, please run the backend service:
-
-```bash
-python deploy_api.py
-```
-
-<br>
-
-## Interface Function Descriptions
-
-### 1. Generate ID Photo (Transparent Background)
-
-Interface Name: `idphoto`
-
-The `Generate ID Photo` interface logic involves sending an RGB image and receiving a standard ID photo and a high-definition ID photo:
-
-- **High-Definition ID Photo**: An ID photo made according to the aspect ratio of `size`, with the filename being `output_image_dir` appended with `_hd` suffix.
-- **Standard ID Photo**: A photo with dimensions equal to `size`, scaled from the high-definition ID photo, with the filename being `output_image_dir`.
-
-It should be noted that both generated photos are transparent (RGBA four-channel images). To produce a complete ID photo, the following `Add Background Color` interface is also required.
-
-> Q: Why is this design used? 
-> A: In actual products, users often need to frequently switch background colors to preview effects. Providing a transparent background image and allowing the front-end JavaScript code to synthesize the color offers a better user experience.
-
-### 2. Add Background Color
-
-Interface Name: `add_background`
-
-The `Add Background Color` interface logic involves sending an RGBA image, adding a background color based on `color`, and synthesizing a JPG image.
-
-### 3. Generate 6-inch Layout Photo
-
-Interface Name: `generate_layout_photos`
-
-The `Generate 6-inch Layout Photo` interface logic involves sending an RGB image (usually an ID photo after adding a background color), arranging the photos according to `size`, and then generating a 6-inch layout photo.
-
-<br>
-
-
-## cURL Request Examples
-
-cURL is a command-line tool used to transfer data using various network protocols. Below are examples of how to use cURL to call these APIs.
-
-### 1. Generate ID Photo (Transparent Background)
-
-```bash
-curl -X POST "http://127.0.0.1:8080/idphoto" \
--F "input_image=@demo/images/test.jpg" \
--F "height=413" \
--F "width=295"
-```
-
-### 2. Add Background Color
-
-```bash
-curl -X POST "http://127.0.0.1:8080/add_background" \
--F "[email protected]" \
--F "color=638cce" \
--F "kb=200"
-```
-
-### 3. Generate Six-Inch Layout Photo
-
-```bash
-curl -X POST "http://127.0.0.1:8080/generate_layout_photos" \
--F "[email protected]" \
--F "height=413" \
--F "width=295" \
--F "kb=200"
-```
-
-## Python Request Example
-
-### 1️⃣ Python Requests Method
-
-#### 1. Generate ID Photo (Transparent Background)
-
-```python
-import requests
-
-url = "http://127.0.0.1:8080/idphoto"
-input_image_path = "images/test.jpg"
-
-files = {"input_image": open(input_image_path, "rb")}
-data = {"height": 413, "width": 295}
-
-response = requests.post(url, files=files, data=data).json()
-
-# response is a JSON dictionary containing status, image_base64_standard, and image_base64_hd
-print(response)
-```
-
-#### 2. Add Background Color
-
-```python
-import requests
-
-url = "http://127.0.0.1:8080/add_background"
-input_image_path = "test.png"
-
-files = {"input_image": open(input_image_path, "rb")}
-data = {"color": '638cce', 'kb': None}
-
-response = requests.post(url, files=files, data=data).json()
-
-# response is a JSON dictionary containing status and image_base64
-print(response)
-```
-
-#### 3. Generate 6-inch Layout Photo
-
-```python
-import requests
-
-url = "http://127.0.0.1:8080/generate_layout_photos"
-input_image_path = "test.jpg"
-
-files = {"input_image": open(input_image_path, "rb")}
-data = {"height": 413, "width": 295, "kb": 200}
-
-response = requests.post(url, files=files, data=data).json()
-
-# response is a JSON dictionary containing status and image_base64
-print(response)
-```
-
-<br>
-
-### 2️⃣ Python Script Request Method
-
-```bash
-python requests_api.py -u <URL> -t <TYPE> -i <INPUT_IMAGE_DIR> -o <OUTPUT_IMAGE_DIR> [--height <HEIGHT>] [--width <WIDTH>] [-c <COLOR>] [-k <KB>]
-```
-
-#### Parameter Descriptions
-
-##### Basic Parameters
-
-- `-u`, `--url`
-
- - **Description**: The URL of the API service.
- - **Default Value**: `http://127.0.0.1:8080`
-
-- `-t`, `--type`
-
- - **Description**: The type of API request, with optional values being `idphoto`, `add_background`, and `generate_layout_photos`. They represent ID photo creation, transparent image background addition, and layout photo generation, respectively.
- - **Default Value**: `idphoto`
-
-- `-i`, `--input_image_dir`
-
- - **Description**: The path of the input image.
- - **Required**: Yes
- - **Example**: `./input_images/photo.jpg`
-
-- `-o`, `--output_image_dir`
- - **Description**: The path to save the image.
- - **Required**: Yes
- - **Example**: `./output_images/processed_photo.jpg`
-
-##### Optional Parameters
-
-- `--height`
-
- - **Description**: The height of the output size for the standard ID photo.
- - **Default Value**: 413
-
-- `--width`
-
- - **Description**: The width of the output size for the standard ID photo.
- - **Default Value**: 295
-
-- `-c`, `--color`
-
- - **Description**: Adds a background color to the transparent image, in Hex format (e.g., #638cce), only effective when the type is `add_background`.
- - **Default Value**: `638cce`
-
-- `-k`, `--kb`
- - **Description**: The KB value of the output photo, only effective when the type is `add_background` or `generate_layout_photos`, and no setting is made when the value is None.
- - **Default Value**: `None`
- - **Example**: `50`
-
-#### 1. Generate ID Photo (Transparent Background)
-
-```bash
-python requests_api.py \
- -u http://127.0.0.1:8080 \
- -t idphoto \
- -i ./photo.jpg \
- -o ./idphoto.png \
- --height 413 \
- --width 295
-```
-
-#### 2. Add Background Color
-
-```bash
-python requests_api.py \
- -u http://127.0.0.1:8080 \
- -t add_background \
- -i ./idphoto.png \
- -o ./idphoto_with_background.jpg \
- -c 638cce \
- -k 50
-```
-
-#### 3. Generate 6-inch Layout Photo
-
-```bash
-python requests_api.py \
- -u http://127.0.0.1:8080 \
- -t generate_layout_photos \
- -i ./idphoto_with_background.jpg \
- -o ./layout_photo.jpg \
- --height 413 \
- --width 295 \
- -k 200
-```
-
-#### Request Failure Scenarios
-
-- The request fails if more than one face is detected in the photo.
-
-## Java Request Example
-
-### Add Maven Dependency
-
-```java
- <dependency>
- <groupId>cn.hutool</groupId>
- <artifactId>hutool-all</artifactId>
- <version>5.8.16</version>
- </dependency>
-
- <dependency>
- <groupId>commons-io</groupId>
- <artifactId>commons-io</artifactId>
- <version>2.6</version>
- </dependency>
-```
-
-### Running the Code
-
-#### 1. Generate ID Photo (Transparent Background)
-
-```java
- /**
- * Generate ID Photo (Transparent Background) /idphoto interface
- * @param inputImageDir File address
- * @return
- * @throws IOException
- */
- public static String requestIdPhoto(String inputImageDir) throws IOException {
- String url = BASE_URL+"/idphoto";
- // Create file object
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("height","413");
- paramMap.put("width","295");
- // Contains status, image_base64_standard, and image_base64_hd
- return HttpUtil.post(url, paramMap);
- }
-```
-
-#### 2. Add Background Color
-
-```java
- /**
- * Add Background Color /add_background interface
- * @param inputImageDir File address
- * @return
- * @throws IOException
- */
- public static String requestAddBackground(String inputImageDir) throws IOException {
- String url = BASE_URL+"/add_background";
- // Create file object
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("color","638cce");
- paramMap.put("kb","200");
- // Response is a JSON dictionary containing status and image_base64
- return HttpUtil.post(url, paramMap);
- }
-```
-
-#### 3. Generate 6-inch Layout Photo
-
-```java
- /**
- * Generate 6-inch Layout Photo /generate_layout_photos interface
- * @param inputImageDir File address
- * @return
- * @throws IOException
- */
- public static String requestGenerateLayoutPhotos(String inputImageDir) throws IOException {
- String url = BASE_URL+"/generate_layout_photos";
- // Create file object
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("height","413");
- paramMap.put("width","295");
- paramMap.put("kb","200");
- // Response is a JSON dictionary containing status and image_base64
- return HttpUtil.post(url, paramMap);
- }
-```
-
-#### 4. Summary
-
-```java
-
-import cn.hutool.http.HttpUtil;
-import cn.hutool.json.JSONObject;
-import cn.hutool.json.JSONUtil;
-import org.apache.commons.io.FileUtils;
-import org.springframework.util.StringUtils;
-import java.io.File;
-import java.io.IOException;
-import java.util.Base64;
-import java.util.HashMap;
-import java.util.Map;
-
-/**
- * @author: qingshuang
- * @createDate: 2024/09/05
- * @description: Java generate ID photo, test case
- */
-public class Test {
- /**
- * Interface address
- */
- private final static String BASE_URL = "http://127.0.0.1:8080";
-
- /**
- * Generate ID Photo (Transparent Background) /idphoto interface
- * @param inputImageDir File address
- * @return
- * @throws IOException
- */
- public static String requestIdPhoto(String inputImageDir) throws IOException {
- String url = BASE_URL+"/idphoto";
- // Create file object
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("height","413");
- paramMap.put("width","295");
- return HttpUtil.post(url, paramMap);
- }
- /**
- * Add Background Color /add_background interface
- * @param inputImageDir File address
- * @return
- * @throws IOException
- */
- public static String requestAddBackground(String inputImageDir) throws IOException {
- String url = BASE_URL+"/add_background";
- // Create file object
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("color","638cce");
- paramMap.put("kb","200");
- return HttpUtil.post(url, paramMap);
- }
- /**
- * Generate 6-inch Layout Photo /generate_layout_photos interface
- * @param inputImageDir File address
- * @return
- * @throws IOException
- */
- public static String requestGenerateLayoutPhotos(String inputImageDir) throws IOException {
- String url = BASE_URL+"/generate_layout_photos";
- // Create file object
- File inputFile = new File(inputImageDir);
- Map<String, Object> paramMap=new HashMap<>();
- paramMap.put("input_image",inputFile);
- paramMap.put("height","413");
- paramMap.put("width","295");
- paramMap.put("kb","200");
- return HttpUtil.post(url, paramMap);
- }
- /**
- * Generate ID Photo (Transparent Background)
- * @param inputImageDir Source file address
- * @param outputImageDir Output file address
- * @throws IOException
- */
- private static void requestIdPhotoToImage(String inputImageDir, String outputImageDir) throws IOException {
- String res =requestIdPhoto(inputImageDir);
- // Convert to JSON
- JSONObject response= JSONUtil.parseObj(res);
- if(response.getBool("status")){// Request interface success
- String image_base64_standard= response.getStr("image_base64_standard");
- String image_base64_hd =response.getStr("image_base64_hd");
- String[] outputImageDirArr= StringUtils.split(outputImageDir,".");
- // Base64 save as image
- FileUtils.writeByteArrayToFile(new File(outputImageDirArr[0]+"_standard."+outputImageDirArr[1]), Base64.getDecoder().decode(image_base64_standard));
- FileUtils.writeByteArrayToFile(new File(outputImageDirArr[0]+"_hd."+outputImageDirArr[1]), Base64.getDecoder().decode(image_base64_hd));
- }
- }
- /**
- * Add Background Color
- * @param inputImageDir Source file address
- * @param outputImageDir Output file address
- * @throws IOException
- */
- private static void requestAddBackgroundToImage(String inputImageDir, String outputImageDir) throws IOException {
- String res =requestAddBackground(inputImageDir);
- // Convert to JSON
- JSONObject response= JSONUtil.parseObj(res);
- if(response.getBool("status")){// Request interface success
- String image_base64= response.getStr("image_base64");
- String[] outputImageDirArr= StringUtils.split(outputImageDir,".");
- // Base64 save as image
- FileUtils.writeByteArrayToFile(new File(outputImageDirArr[0]+"_background."+outputImageDirArr[1]), Base64.getDecoder().decode(image_base64));
- }
- }
- /**
- * Generate 6-inch Layout Photo
- * @param inputImageDir Source file address
- * @param outputImageDir Output file address
- * @throws IOException
- */
- private static void requestGenerateLayoutPhotosToImage(String inputImageDir, String outputImageDir) throws IOException {
- String res =requestGenerateLayoutPhotos(inputImageDir);
- // Convert to JSON
- JSONObject response= JSONUtil.parseObj(res);
- if(response.getBool("status")){// Request interface success
- String image_base64= response.getStr("image_base64");
- String[] outputImageDirArr= StringUtils.split(outputImageDir,".");
- // Base64 save as image
- FileUtils.writeByteArrayToFile(new File(outputImageDirArr[0]+"_layout."+outputImageDirArr[1]), Base64.getDecoder().decode(image_base64));
- }
- }
-
- public static void main(String[] args) {
- try {
- // Generate ID Photo (Transparent Background)
- requestIdPhotoToImage("C:\\Users\\Administrator\\Desktop\\1111.jpg","C:\\Users\\Administrator\\Desktop\\2222.png");
- // Add Background Color
- requestAddBackgroundToImage("C:\\Users\\Administrator\\Desktop\\2222_hd.png","C:\\Users\\Administrator\\Desktop\\idphoto_with_background.jpg");
- // Generate 6-inch Layout Photo
- requestGenerateLayoutPhotosToImage("C:\\Users\\Administrator\\Desktop\\1111.jpg","C:\\Users\\Administrator\\Desktop\\2222.png");
-
- } catch (IOException e) {
- e.printStackTrace();
- }
- }
-}
-```
-
-## JavaScript Request Examples
-
-In JavaScript, we can use the `fetch` API to send HTTP requests. Below are examples of how to call these APIs using JavaScript.
-
-### 1. Generate ID Photo (Transparent Background)
-
-```javascript
-async function generateIdPhoto(inputImagePath, height, width) {
- const url = "http://127.0.0.1:8080/idphoto";
- const formData = new FormData();
- formData.append("input_image", new File([await fetch(inputImagePath).then(res => res.blob())], "test.jpg"));
- formData.append("height", height);
- formData.append("width", width);
-
- const response = await fetch(url, {
- method: 'POST',
- body: formData
- });
-
- const result = await response.json();
- console.log(result);
- return result;
-}
-
-// Example call
-generateIdPhoto("images/test.jpg", 413, 295).then(response => {
- console.log(response);
-});
-```
-
-### 2. Add Background Color
-
-```javascript
-async function addBackground(inputImagePath, color, kb) {
- const url = "http://127.0.0.1:8080/add_background";
- const formData = new FormData();
- formData.append("input_image", new File([await fetch(inputImagePath).then(res => res.blob())], "test.png"));
- formData.append("color", color);
- formData.append("kb", kb);
-
- const response = await fetch(url, {
- method: 'POST',
- body: formData
- });
-
- const result = await response.json();
- console.log(result);
- return result;
-}
-
-// Example call
-addBackground("test.png", "638cce", 200).then(response => {
- console.log(response);
-});
-```
-
-### 3. Generate Six-Inch Layout Photo
-
-```javascript
-async function generateLayoutPhotos(inputImagePath, height, width, kb) {
- const url = "http://127.0.0.1:8080/generate_layout_photos";
- const formData = new FormData();
- formData.append("input_image", new File([await fetch(inputImagePath).then(res => res.blob())], "test.jpg"));
- formData.append("height", height);
- formData.append("width", width);
- formData.append("kb", kb);
-
- const response = await fetch(url, {
- method: 'POST',
- body: formData
- });
-
- const result = await response.json();
- console.log(result);
- return result;
-}
-
-// Example call
-generateLayoutPhotos("test.jpg", 413, 295, 200).then(response => {
- console.log(response);
-});
-```

hivision/creator/__init__.py

@@ -12,7 +12,7 @@ from typing import Tuple
 import hivision.creator.utils as U
 from .context import Context, ContextHandler, Params, Result
 from .human_matting import extract_human
-from .face_detector import detect_face
+from .face_detector import detect_face_mtcnn, detect_face_face_plusplus
 from .photo_adjuster import adjust_photo
 
 
@@ -41,7 +41,9 @@ class IDCreator:
  """
  # 处理者
  self.matting_handler: ContextHandler = extract_human
- self.detection_handler: ContextHandler = detect_face
+ self.detection_handler: ContextHandler = detect_face_face_plusplus
+ # self.detection_handler: ContextHandler = detect_face_mtcnn
+
  # 上下文
  self.ctx = None
 

hivision/creator/face_detector.py

@@ -9,15 +9,19 @@ r"""
 """
 from mtcnnruntime import MTCNN
 from .context import Context
-from hivision.error import FaceError
+from hivision.error import FaceError, APIError
+from hivision.utils import resize_image_to_kb_base64
+import requests
 import cv2
+import os
+
 
 mtcnn = None
 
 
-def detect_face(ctx: Context, scale: int = 2):
+def detect_face_mtcnn(ctx: Context, scale: int = 2):
  """
- 人脸检测处理者，只进行人脸数量的检测
+ 基于MTCNN模型的人脸检测处理器，只进行人脸数量的检测
  :param ctx: 上下文，此时已获取到原始图和抠图结果，但是我们只需要原始图
  :param scale: 最大边长缩放比例，原图:缩放图 = 1:scale
  :raise FaceError: 人脸检测错误，多个人脸或者没有人脸
@@ -30,13 +34,91 @@ def detect_face(ctx: Context, scale: int = 2):
  (ctx.origin_image.shape[1] // scale, ctx.origin_image.shape[0] // scale),
  interpolation=cv2.INTER_AREA,
  )
- faces, _ = mtcnn.detect(image)
+ faces, _ = mtcnn.detect(image, thresholds=[0.8, 0.8, 0.8])
+ # print(len(faces))
  if len(faces) != 1:
  # 保险措施，如果检测到多个人脸或者没有人脸，用原图再检测一次
  faces, _ = mtcnn.detect(ctx.origin_image)
  else:
+ # 如果只有一个人脸，将人脸坐标放大
  for item, param in enumerate(faces[0]):
  faces[0][item] = param * 2
  if len(faces) != 1:
  raise FaceError("Expected 1 face, but got {}".format(len(faces)), len(faces))
- ctx.face = (faces[0][0], faces[0][1], faces[0][2], faces[0][3], faces[0][4])
+
+ left = faces[0][0]
+ top = faces[0][1]
+ width = faces[0][2] - left + 1
+ height = faces[0][3] - top + 1
+
+ ctx.face = (left, top, width, height)
+
+
+def detect_face_face_plusplus(ctx: Context):
+ """
+ 基于Face++ API接口的人脸检测处理器，只进行人脸数量的检测
+ :param ctx: 上下文，此时已获取到原始图和抠图结果，但是我们只需要原始图
+ :param scale: 最大边长缩放比例，原图:缩放图 = 1:scale
+ :raise FaceError: 人脸检测错误，多个人脸或者没有人脸
+ """
+ url = "https://api-cn.faceplusplus.com/facepp/v3/detect"
+ api_key = os.getenv("FACE_PLUS_API_KEY")
+ api_secret = os.getenv("FACE_PLUS_API_SECRET")
+
+ image = ctx.origin_image
+ # 将图片转为 base64, 且不大于2MB（Face++ API接口限制）
+ image_base64 = resize_image_to_kb_base64(image, 2000, mode="max")
+
+ files = {
+ "api_key": (None, api_key),
+ "api_secret": (None, api_secret),
+ "image_base64": (None, image_base64),
+ }
+
+ # 发送 POST 请求
+ response = requests.post(url, files=files)
+
+ # 获取响应状态码
+ status_code = response.status_code
+ response_json = response.json()
+
+ if status_code == 200:
+ face_num = response_json["face_num"]
+ if face_num == 1:
+ face_rectangle = response_json["faces"][0]["face_rectangle"]
+ ctx.face = (
+ face_rectangle["left"],
+ face_rectangle["top"],
+ face_rectangle["width"],
+ face_rectangle["height"],
+ )
+ else:
+ raise FaceError(
+ "Expected 1 face, but got {}".format(face_num), len(face_num)
+ )
+
+ elif status_code == 401:
+ raise APIError(
+ f"Face++ Status code {status_code} Authentication error: API key and secret do not match.",
+ status_code,
+ )
+
+ elif status_code == 403:
+ reason = response_json.get("error_message", "Unknown authorization error.")
+ raise APIError(
+ f"Authorization error: {reason}",
+ status_code,
+ )
+
+ elif status_code == 400:
+ error_message = response_json.get("error_message", "Bad request.")
+ raise APIError(
+ f"Bad request error: {error_message}",
+ status_code,
+ )
+
+ elif status_code == 413:
+ raise APIError(
+ f"Face++ Status code {status_code} Request entity too large: The image exceeds the 2MB limit.",
+ status_code,
+ )

hivision/creator/photo_adjuster.py

@@ -21,7 +21,7 @@ def adjust_photo(ctx: Context):
  standard_size = ctx.params.size
  params = ctx.params
  x, y = face_rect[0], face_rect[1]
- w, h = face_rect[2] - x + 1, face_rect[3] - y + 1
+ w, h = face_rect[2], face_rect[3]
  height, width = ctx.processing_image.shape[:2]
  width_height_ratio = standard_size[0] / standard_size[1]
  # Step2. 计算高级参数

hivision/error.py

@@ -19,3 +19,15 @@ class FaceError(Exception):
  """
  super().__init__(err)
  self.face_num = face_num
+
+
+class APIError(Exception):
+ def __init__(self, err, status_code):
+ """
+ API错误
+ Args:
+ err: 错误描述
+ status_code: 告诉此时的错误状态码
+ """
+ super().__init__(err)
+ self.status_code = status_code

hivision/utils.py

@@ -74,13 +74,20 @@ def resize_image_to_kb(input_image, output_image_path, target_size_kb):
  quality = 1
 
 
-def resize_image_to_kb_base64(input_image, target_size_kb):
+import numpy as np
+from PIL import Image
+import io
+import base64
+
+
+def resize_image_to_kb_base64(input_image, target_size_kb, mode="exact"):
  """
  Resize an image to a target size in KB and return it as a base64 encoded string.
  将图像调整大小至目标文件大小（KB）并返回base64编码的字符串。
 
  :param input_image: Input image as a NumPy array or PIL Image. 输入图像，可以是NumPy数组或PIL图像。
  :param target_size_kb: Target size in KB. 目标文件大小（KB）。
+ :param mode: Mode of resizing ('exact', 'max', 'min'). 模式：'exact'（精确大小）、'max'（不大于）、'min'（不小于）。
 
  :return: Base64 encoded string of the resized image. 调整大小后的图像的base64编码字符串。
  """
@@ -109,19 +116,30 @@ def resize_image_to_kb_base64(input_image, target_size_kb):
  # Get the size of the image in KB
  img_size_kb = len(img_byte_arr.getvalue()) / 1024
 
- # Check if the image size is within the target size
- if img_size_kb <= target_size_kb or quality == 1:
+ # Check based on the mode
+ if mode == "exact":
+ # If the image size is equal to the target size, we can return it
+ if img_size_kb == target_size_kb:
+ break
+
  # If the image is smaller than the target size, add padding
- if img_size_kb < target_size_kb:
+ elif img_size_kb < target_size_kb:
  padding_size = int(
  (target_size_kb * 1024) - len(img_byte_arr.getvalue())
  )
  padding = b"\x00" * padding_size
  img_byte_arr.write(padding)
+ break
+
+ elif mode == "max":
+ # If the image size is within the target size, we can return it
+ if img_size_kb <= target_size_kb or quality == 1:
+ break
 
- # Encode the image data to base64
- img_base64 = base64.b64encode(img_byte_arr.getvalue()).decode("utf-8")
- return img_base64
+ elif mode == "min":
+ # If the image size is greater than or equal to the target size, we can return it
+ if img_size_kb >= target_size_kb:
+ break
 
  # Reduce the quality if the image is still too large
  quality -= 5
@@ -130,6 +148,10 @@ def resize_image_to_kb_base64(input_image, target_size_kb):
  if quality < 1:
  quality = 1
 
+ # Encode the image data to base64
+ img_base64 = base64.b64encode(img_byte_arr.getvalue()).decode("utf-8")
+ return img_base64
+
 
 def numpy_2_base64(img: np.ndarray):
  _, buffer = cv2.imencode(".png", img)

inference.py

@@ -1,107 +0,0 @@
-import os
-import cv2
-import argparse
-import numpy as np
-import onnxruntime
-from hivision.error import FaceError
-from hivision.utils import hex_to_rgb, resize_image_to_kb, add_background
-from hivision import IDCreator
-from hivision.creator.layout_calculator import (
- generate_layout_photo,
- generate_layout_image,
-)
-
-parser = argparse.ArgumentParser(description="HivisionIDPhotos 证件照制作推理程序。")
-
-creator = IDCreator()
-
-parser.add_argument(
- "-t",
- "--type",
- help="请求 API 的种类，有 idphoto、add_background 和 generate_layout_photos 可选",
- default="idphoto",
-)
-parser.add_argument("-i", "--input_image_dir", help="输入图像路径", required=True)
-parser.add_argument("-o", "--output_image_dir", help="保存图像路径", required=True)
-parser.add_argument("--height", help="证件照尺寸-高", default=413)
-parser.add_argument("--width", help="证件照尺寸-宽", default=295)
-parser.add_argument("-c", "--color", help="证件照背景色", default="638cce")
-parser.add_argument(
- "-k", "--kb", help="输出照片的 KB 值，仅对换底和制作排版照生效", default=None
-)
-
-args = parser.parse_args()
-
-root_dir = os.path.dirname(os.path.abspath(__file__))
-
-# 预加载 ONNX 模型
-print("正在加载抠图模型...")
-# HY_HUMAN_MATTING_WEIGHTS_PATH = os.path.join(
-# root_dir, "hivision/creator/weights/hivision_modnet.onnx"
-# )
-# sess = onnxruntime.InferenceSession(HY_HUMAN_MATTING_WEIGHTS_PATH)
-
-input_image = cv2.imread(args.input_image_dir, cv2.IMREAD_UNCHANGED)
-
-
-# 如果模式是生成证件照
-if args.type == "idphoto":
-
- # 将字符串转为元组
- size = (int(args.height), int(args.width))
- try:
- result = creator(input_image, size=size)
- except FaceError:
- print("人脸数量不等于 1，请上传单张人脸的图像。")
- else:
- # 保存标准照
- cv2.imwrite(args.output_image_dir, result.standard)
-
- # 保存高清照
- file_name, file_extension = os.path.splitext(args.output_image_dir)
- new_file_name = file_name + "_hd" + file_extension
- cv2.imwrite(new_file_name, result.hd)
-
-# 如果模式是添加背景
-elif args.type == "add_background":
-
- # 将字符串转为元组
- color = hex_to_rgb(args.color)
- # 将元祖的 0 和 2 号数字交换
- color = (color[2], color[1], color[0])
-
- result_image = add_background(input_image, bgr=color)
- result_image = result_image.astype(np.uint8)
-
- if args.kb:
- result_image = cv2.cvtColor(result_image, cv2.COLOR_RGB2BGR)
- result_image = resize_image_to_kb(
- result_image, args.output_image_dir, int(args.kb)
- )
- else:
- cv2.imwrite(args.output_image_dir, result_image)
-
-# 如果模式是生成排版照
-elif args.type == "generate_layout_photos":
-
- size = (int(args.height), int(args.width))
-
- typography_arr, typography_rotate = generate_layout_photo(
- input_height=size[0], input_width=size[1]
- )
-
- result_layout_image = generate_layout_image(
- input_image,
- typography_arr,
- typography_rotate,
- height=size[0],
- width=size[1],
- )
-
- if args.kb:
- result_layout_image = cv2.cvtColor(result_layout_image, cv2.COLOR_RGB2BGR)
- result_layout_image = resize_image_to_kb(
- result_layout_image, args.output_image_dir, int(args.kb)
- )
- else:
- cv2.imwrite(args.output_image_dir, result_layout_image)

requests_api.py

@@ -1,119 +0,0 @@
-import requests
-import base64
-import argparse
-import os
-
-
-def base64_save(_base64_image_data, save_path):
- # 解码 Base64 数据并保存为 PNG 文件
- img_data = base64.b64decode(_base64_image_data)
- with open(save_path, "wb") as file:
- file.write(img_data)
-
-
-# 读取本地图像文件并转换为Base64编码
-def file_2_base64(file_path):
- with open(file_path, "rb") as file:
- encoded_string = base64.b64encode(file.read()).decode("utf-8")
- return encoded_string
-
-
-# 发送请求到 /idphoto 接口
-def request_idphoto(file_path, height, width):
- files = {"input_image": open(file_path, "rb")}
- data = {"height": int(height), "width": int(width)}
- response = requests.post(url, files=files, data=data)
- return response.json()
-
-
-# 发送请求到 /add_background 接口
-def request_add_background(file_path, color, kb=None):
- files = {"input_image": open(file_path, "rb")}
- data = {"color": str(color), "kb": kb}
- response = requests.post(url, files=files, data=data)
- return response.json()
-
-
-# 发送请求到 /generate_layout_photos 接口
-def request_generate_layout_photos(file_path, height, width, kb=None):
- files = {"input_image": open(file_path, "rb")}
- data = {"height": height, "width": width, "kb": kb}
- response = requests.post(url, files=files, data=data)
- return response.json()
-
-
-# 示例调用
-if __name__ == "__main__":
-
- parser = argparse.ArgumentParser(
- description="HivisionIDPhotos 证件照制作推理程序。"
- )
- parser.add_argument(
- "-u", "--url", help="API 服务的 URL", default="http://localhost:8080"
- )
-
- parser.add_argument(
- "-t",
- "--type",
- help="请求 API 的种类，有 idphoto、add_background 和 generate_layout_photos 可选",
- default="idphoto",
- )
- parser.add_argument("-i", "--input_image_dir", help="输入图像路径", required=True)
- parser.add_argument("-o", "--output_image_dir", help="保存图像路径", required=True)
- parser.add_argument("--height", help="证件照尺寸-高", default=413)
- parser.add_argument("--width", help="证件照尺寸-宽", default=295)
- parser.add_argument("-c", "--color", help="证件照背景色", default="638cce")
- parser.add_argument(
- "-k", "--kb", help="输出照片的 KB 值，仅对换底和制作排版照生效", default=None
- )
- args = parser.parse_args()
-
- url = f"{args.url}/{args.type}" # 替换为实际的接口 URL
- # color = hex_to_rgb(args.color)
- # color = (color[2], color[1], color[0])
-
- if args.type == "idphoto":
- # 调用 /idphoto 接口
- idphoto_response = request_idphoto(
- args.input_image_dir, int(args.height), int(args.width)
- )
-
- if idphoto_response["status"]:
- # 解码 Base64 数据并保存为 PNG 文件
- base64_image_data_standard = idphoto_response["image_base64_standard"]
- base64_image_data_standard_hd = idphoto_response["image_base64_hd"]
-
- file_name, file_extension = os.path.splitext(args.output_image_dir)
- # 定义新的文件路径（在原有的文件名后添加"_hd"）
- new_file_name = file_name + "_hd" + file_extension
-
- # 解码 Base64 数据并保存为 PNG 文件
- base64_save(base64_image_data_standard, args.output_image_dir)
- base64_save(base64_image_data_standard_hd, new_file_name)
-
- print(f"请求{args.type}接口成功，已保存图像。")
- else:
- print("人脸数量不等于 1，请上传单张人脸的图像。")
-
- elif args.type == "add_background":
- # 调用 /add_background 接口
- add_background_response = request_add_background(
- args.input_image_dir, args.color, kb=args.kb
- )
- base64_image_data = add_background_response["image_base64"]
- base64_save(base64_image_data, args.output_image_dir)
-
- print(f"请求{args.type}接口成功，已保存图像。")
-
- elif args.type == "generate_layout_photos":
- # 调用 /generate_layout_photos 接口
- generate_layout_response = request_generate_layout_photos(
- args.input_image_dir, int(args.height), int(args.width), args.kb
- )
- base64_image_data = generate_layout_response["image_base64"]
- base64_save(base64_image_data, args.output_image_dir)
-
- print(f"请求{args.type}接口成功，已保存图像。")
-
- else:
- print("不支持的 API 类型，请检查输入参数。")

requirements-app.txt

@@ -1,2 +0,0 @@
-gradio>=4.43.0
-fastapi

requirements.txt

@@ -3,4 +3,5 @@ onnxruntime>=1.15.0
 numpy<=1.26.4
 mtcnn-runtime
 gradio>=4.43.0
-fastapi
+fastapi
+requests

scripts/build_pypi.py

@@ -1,9 +0,0 @@
-#!/usr/bin/env python
-# -*- coding: utf-8 -*-
-r"""
-@DATE: 2024/9/5 16:56
-@File: build_pypi.py
-@IDE: pycharm
-@Description:
- 构建pypi包
-"""

test/create_id_photo.py

@@ -1,22 +0,0 @@
-#!/usr/bin/env python
-# -*- coding: utf-8 -*-
-r"""
-@DATE: 2024/9/5 21:39
-@File: create_id_photo.py
-@IDE: pycharm
-@Description:
- 用于测试创建证件照
-"""
-from hivision.creator import IDCreator
-import cv2
-import os
-
-now_dir = os.path.dirname(__file__)
-image_path = os.path.join(os.path.dirname(now_dir), "app", "images", "test.jpg")
-output_dir = os.path.join(now_dir, "temp")
-
-image = cv2.imread(image_path)
-creator = IDCreator()
-result = creator(image)
-cv2.imwrite(os.path.join(output_dir, "result.png"), result.standard)
-cv2.imwrite(os.path.join(output_dir, "result_hd.png"), result.hd)

test/temp/.gitkeep

@@ -1 +0,0 @@
-存放一些测试临时文件