人体分割常用语言和常见情况的示例代码_视觉智能开放平台(VIAPI)-阿里云帮助中心

本文档为您介绍人体分割常用语言和常见情况的示例代码。

说明

您可以进入在线咨询获取在线人工帮助。
阿里云视觉智能开放平台视觉AI能力API接入、接口使用或问题咨询等，请通过钉钉群（23109592）加入阿里云视觉智能开放平台咨询群联系我们。

能力介绍

关于人体分割的功能介绍以及具体调用参数说明，请参见人体分割。

SDK包安装

常见语言的SDK依赖包信息，请参见SDK总览。

配置环境变量

配置环境变量ALIBABA_CLOUD_ACCESS_KEY_ID和ALIBABA_CLOUD_ACCESS_KEY_SECRET。

重要

阿里云账号AccessKey拥有所有API的访问权限，建议您使用RAM用户进行API访问或日常运维，具体操作，请参见创建RAM用户。
请不要将AccessKey ID和AccessKey Secret保存到工程代码里，否则可能导致AccessKey泄露，威胁您账号下所有资源的安全。

Linux和macOS系统配置方法
1. 在IntelliJ IDEA中打开终端Terminal。
2. 执行以下命令，配置环境变量。
  <access_key_id>需替换为您RAM用户的AccessKey ID，<access_key_secret>替换为您RAM用户的AccessKey Secret。如果后续需要进行更多权限相关的配置，具体操作请参见使用RAM Policy控制访问权限。
```
export ALIBABA_CLOUD_ACCESS_KEY_ID=<access_key_id> 
export ALIBABA_CLOUD_ACCESS_KEY_SECRET=<access_key_secret>
```
Windows系统配置方法
新建环境变量文件，添加环境变量ALIBABA_CLOUD_ACCESS_KEY_ID和ALIBABA_CLOUD_ACCESS_KEY_SECRET，并写入已准备好的AccessKey ID和AccessKey Secret。然后重启Windows系统。本操作以Windows 10为例进行说明。
1. 打开文件资源管理器，在此电脑上右键单击属性。
2. 在右侧导航栏，单击高级系统配置。
3. 在系统属性对话框的高级页签下，单击环境变量。
4. 在环境变量对话框中，单击新建(W)。
5. 在弹出的新建系统变量对话框中，添加环境变量ALIBABA_CLOUD_ACCESS_KEY_ID和ALIBABA_CLOUD_ACCESS_KEY_SECRET，并写入已准备好的AccessKey ID和AccessKey Secret。
6. 重启Windows系统，使配置生效。

示例代码

文件在上海地域OSS

Java

/*
引入依赖包
<!-- https://mvnrepository.com/artifact/com.aliyun/imageseg20191230 -->
<dependency>
      <groupId>com.aliyun</groupId>
      <artifactId>imageseg20191230</artifactId>
      <version>${aliyun.imageseg.version}</version>
</dependency>
*/

import com.aliyun.imageseg20191230.models.SegmentBodyResponse;
import com.aliyun.tea.TeaException;
import com.aliyun.tea.TeaModel;

public class SegmentBody {
    public static com.aliyun.imageseg20191230.Client createClient(String accessKeyId, String accessKeySecret) throws Exception {
        /*
          初始化配置对象com.aliyun.teaopenapi.models.Config
          Config对象存放 AccessKeyId、AccessKeySecret、endpoint等配置
         */
         com.aliyun.teaopenapi.models.Config config = new com.aliyun.teaopenapi.models.Config()
                .setAccessKeyId(accessKeyId)
                .setAccessKeySecret(accessKeySecret);
        // 访问的域名
        config.endpoint = "imageseg.cn-shanghai.aliyuncs.com";
        return new com.aliyun.imageseg20191230.Client(config);
    }

    public static void main(String[] args_) throws Exception {
        // 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html
        // 如果您使用的是RAM用户的AccessKey，还需要为子账号授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html
        // 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行代码示例前必须先配置环境变量。
        String accessKeyId = System.getenv("ALIBABA_CLOUD_ACCESS_KEY_ID");
        String accessKeySecret = System.getenv("ALIBABA_CLOUD_ACCESS_KEY_SECRET"); 
        com.aliyun.imageseg20191230.Client client = SegmentBody.createClient(accessKeyId, accessKeySecret);
        com.aliyun.imageseg20191230.models.SegmentBodyRequest segmentBodyRequest = new com.aliyun.imageseg20191230.models.SegmentBodyRequest()
                .setImageURL("http://viapi-test.oss-cn-shanghai.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody1.png");
        com.aliyun.teautil.models.RuntimeOptions runtime = new com.aliyun.teautil.models.RuntimeOptions();
        try {
            SegmentBodyResponse segmentBodyResponse = client.segmentBodyWithOptions(segmentBodyRequest, runtime);
            // 获取整体结果
            System.out.println(com.aliyun.teautil.Common.toJSONString(TeaModel.buildMap(segmentBodyResponse)));
            // 获取单个字段
            System.out.println(segmentBodyResponse.getBody().getData().getImageURL());
        } catch (TeaException teaException) {
            // 获取整体报错信息
            System.out.println(com.aliyun.teautil.Common.toJSONString(teaException));
            // 获取单个字段
            System.out.println(teaException.getCode());
        }
    }
}

Python

# -*- coding: utf-8 -*-
# 引入依赖包
# pip install alibabacloud_imageseg20191230

import os
from alibabacloud_imageseg20191230.client import Client
from alibabacloud_imageseg20191230.models import SegmentBodyRequest
from alibabacloud_tea_openapi.models import Config
from alibabacloud_tea_util.models import RuntimeOptions

# 初始化Config
config = Config(
    # 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html。
    # 如果您用的是RAM用户的AccessKey，还需要为RAM用户授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html。
    # 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行代码示例前必须先配置环境变量。
    access_key_id=os.environ.get('ALIBABA_CLOUD_ACCESS_KEY_ID'),
    access_key_secret=os.environ.get('ALIBABA_CLOUD_ACCESS_KEY_SECRET'),
    # 访问的域名。
    endpoint='imageseg.cn-shanghai.aliyuncs.com',
    # 访问的域名对应的region
    region_id='cn-shanghai'
)
request = SegmentBodyRequest(
    image_url='https://viapi-test.oss-cn-shanghai.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody_IDCard.jpeg',
    #ReturnForm参数设置：
    #指定返回的图像形式。
    #如果设置为mask，则返回单通道黑白图。
    #如果设置为crop，则返回裁剪之后的四通道PNG图（裁掉边缘空白区域）。
    #如果设置为whiteBK，则返回白底图。
    #如果不设置或者设置为其他值，则返回四通道PNG图。
    return_form='mask'
)
# 初始化RuntimeObject
runtime_option = RuntimeOptions()
try:
  # 初始化Client
  client = Client(config)
  response = client.segment_body_with_options(request, runtime_option)
  # 获取整体结果
  print(response.body)
except Exception as error:
  # 获取整体报错信息
  print(error)
  # 获取单个字段
  print(error.code)
  # tips: 可通过error.__dict__查看属性名称

PHP

<?php

//安装依赖包
//composer require alibabacloud/imageseg-20191230

use AlibabaCloud\SDK\Imageseg\V20191230\Imageseg;
use AlibabaCloud\Tea\Utils\Utils;
use \Exception;
use Darabonba\OpenApi\Models\Config;
use AlibabaCloud\SDK\Imageseg\V20191230\Models\SegmentBodyRequest;
use AlibabaCloud\Tea\Utils\Utils\RuntimeOptions;

class SegmentBody {

    /**
     * 使用AK&SK初始化账号Client
     * @param string $accessKeyId
     * @param string $accessKeySecret
     * @return Imageseg Client
     */
    public static function createClient($accessKeyId, $accessKeySecret){
        //初始化配置对象Darabonba\OpenApi\Models\Config。
        //Config对象存放accessKeyId、accessKeySecret、endpoint等配置
        $config = new Config([
            "accessKeyId" => $accessKeyId,
            "accessKeySecret" => $accessKeySecret
        ]);
        // 访问的域名
        $config->endpoint = "imageseg.cn-shanghai.aliyuncs.com";
        return new Imageseg($config);
    }
    /**
     * @param string[] $args
     * @return void
     */
    public static function main($args){
        // 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html。
        // 如果您用的是RAM用户的AccessKey，还需要为RAM用户授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html。
        // 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行示例前必须先配置环境变量。    
        $accessKeyId = getenv('ALIBABA_CLOUD_ACCESS_KEY_ID');      
        $accessKeySecret = getenv('ALIBABA_CLOUD_ACCESS_KEY_SECRET');
        $client = self::createClient($accessKeyId, $accessKeySecret)
        $segmentBodyRequest = new SegmentBodyRequest([
            "imageURL" => "http://viapi-test.oss-cn-shanghai.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody9.png",
            "returnForm" => "mask"
        ]);
        $runtime = new RuntimeOptions([]);
        try {
            $resp = $client->segmentBodyWithOptions($segmentBodyRequest, $runtime);
            # 获取整体结果
            echo Utils::toJSONString($resp->body);
        } catch (Exception $exception) {
            # 获取整体报错信息
            echo Utils::toJSONString($exception);
            # 获取单个字段
            echo $exception->getCode();
        }
    }
}
$path = __DIR__ . \DIRECTORY_SEPARATOR . '..' . \DIRECTORY_SEPARATOR . 'vendor' . \DIRECTORY_SEPARATOR . 'autoload.php';
if (file_exists($path)) {
    require_once $path;
}
//$argv是预留的数组入参参数，无实际意义，无需进行修改
SegmentBody::main(array_slice($argv, 1));

Node.js

// 安装依赖包
// npm install @alicloud/imageseg20191230
const ImagesegClient = require('@alicloud/imageseg20191230');
const OpenapiClient = require('@alicloud/openapi-client');
const TeaUtil = require('@alicloud/tea-util');

let config = new OpenapiClient.Config({
    // 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html。
    // 如果您用的是RAM用户AccessKey，还需要为RAM用户授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html。
    // 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行示例前必须先配置环境变量。 
    accessKeyId: process.env.ALIBABA_CLOUD_ACCESS_KEY_ID,   
    accessKeySecret: process.env.ALIBABA_CLOUD_ACCESS_KEY_SECRET
});
// 访问的域名
config.endpoint = `imageseg.cn-shanghai.aliyuncs.com`;
const client = new ImagesegClient.default(config);
let segmentBodyRequest = new ImagesegClient.SegmentBodyRequest({
  imageURL: "http://viapi-test.oss-cn-shanghai.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody1.png",
});
let runtime = new TeaUtil.RuntimeOptions({ });
client.segmentBodyWithOptions(segmentBodyRequest, runtime)
  .then(function(segmentBodyRequestResponse) {
    // 获取整体结果
    console.log(segmentBodyRequestResponse);
    // 获取单个字段
    console.log(segmentBodyRequestResponse.body.data);
  }, function(error) {
    // 获取整体报错信息
    console.log(error);
    // 获取单个字段
    console.log(error.data.Code);
  })

/**
依赖github.com/alibabacloud-go/imageseg-20191230/v2
建议使用go mod tidy安装依赖
 */

import (
	"fmt"
	openapi "github.com/alibabacloud-go/darabonba-openapi/v2/client"
	imageseg20191230 "github.com/alibabacloud-go/imageseg-20191230/v2/client"
	util "github.com/alibabacloud-go/tea-utils/v2/service"
	"github.com/alibabacloud-go/tea/tea"
)

func main() {
  // 从环境变量读取配置的AccessKey ID
  // 运行示例前必须先配置环境变量  
  // 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html。
  // 如果您用的是RAM用户的AccessKey，还需要为RAM用户授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html。
  // 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行示例前必须先配置环境变量。  
  accessKeyId := os.Getenv("ALIBABA_CLOUD_ACCESS_KEY_ID")  
  accessKeySecret := os.Getenv("ALIBABA_CLOUD_ACCESS_KEY_SECRET")
  // 初始化配置对象 &openapi.Config。Config对象存放AccessKeyId、AccessKeySecret、Endpoint等配置。
  config := &openapi.Config{
    AccessKeyId: tea.String(accessKeyId),
    AccessKeySecret: tea.String(accessKeySecret)
	}
	// 访问的域名
	config.Endpoint = tea.String("imageseg.cn-shanghai.aliyuncs.com")
	client, err := imageseg20191230.NewClient(config)
	if err != nil {
		panic(err)
	}
	segmentBodyRequest := &imageseg20191230.SegmentBodyRequest{
		ImageURL: tea.String("http://viapi-test.oss-cn-shanghai.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody1.png"),
	}
	runtime := &util.RuntimeOptions{}
	segmentBodyResponse, err := client.SegmentBodyWithOptions(segmentBodyRequest, runtime)
	if err != nil {
		// 获取整体报错信息
		fmt.Println(err.Error())
	} else {
		// 获取整体结果
		fmt.Println(segmentBodyResponse)
	}
}

// 安装依赖包
// dotnet add package AlibabaCloud.SDK.Imageseg20191230
using System;
using System.Collections;
using System.Collections.Generic;
using System.IO;
using System.Threading.Tasks;
using AlibabaCloud.SDK.Imageseg20191230.Models;
using Tea;
using Tea.Utils;

namespace AlibabaCloud.SDK.Sample
{
    public class Sample
    {
      /**
          * 使用AK&SK初始化账号Client
          * @param accessKeyId
          * @param accessKeySecret
          * @return Client
          * @throws Exception
      */
        public static AlibabaCloud.SDK.Imageseg20191230.Client CreateClient(string accessKeyId, string accessKeySecret)
        {
            AlibabaCloud.OpenApiClient.Models.Config config = new AlibabaCloud.OpenApiClient.Models.Config
            {
              AccessKeyId = accessKeyId,
              AccessKeySecret = accessKeySecret,
            };
            // 访问的域名
            config.Endpoint = "imageseg.cn-shanghai.aliyuncs.com";
            return new AlibabaCloud.SDK.Imageseg20191230.Client(config);
        }
        public static void Main(string[] args)
        {
            // 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html
            // 如果您使用的是RAM用户的AccessKey，还需要为子账号授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html
            // 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行代码示例前必须先配置环境变量。
            AlibabaCloud.SDK.Imageseg20191230.Client client = CreateClient(System.Environment.GetEnvironmentVariable("ALIBABA_CLOUD_ACCESS_KEY_ID"), System.Environment.GetEnvironmentVariable("ALIBABA_CLOUD_ACCESS_KEY_SECRET"));
            AlibabaCloud.SDK.Imageseg20191230.Models.SegmentBodyRequest segmentBodyRequest = new AlibabaCloud.SDK.Imageseg20191230.Models.SegmentBodyRequest
            {
                ImageURL = "http://viapi-test.oss-cn-shanghai.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody1.png",
            };
            AlibabaCloud.TeaUtil.Models.RuntimeOptions runtime = new AlibabaCloud.TeaUtil.Models.RuntimeOptions();
            try
            {
                AlibabaCloud.SDK.Imageseg20191230.Models.SegmentBodyResponse segmentBodyResponse = client.SegmentBodyWithOptions(segmentBodyRequest, runtime);
                // 获取整体结果
                Console.WriteLine(AlibabaCloud.TeaUtil.Common.ToJSONString(segmentBodyResponse.Body));
                // 获取单个字段
                Console.WriteLine(AlibabaCloud.TeaUtil.Common.ToJSONString(segmentBodyResponse.Body.Data));
            }
            catch (TeaException error)
            {
                // 如有需要，请打印 error
                Console.WriteLine(AlibabaCloud.TeaUtil.Common.AssertAsString(error.Message));
            }
            catch (Exception _error)
            {
                TeaException error = new TeaException(new Dictionary<string, object>
              {
                { "message", _error.Message }
              });
                // 如有需要，请打印 error
                Console.WriteLine(AlibabaCloud.TeaUtil.Common.AssertAsString(error.Message));
            }
        }
    }
}

文件在本地或可访问的URL

Java

/*
引入依赖包
<!-- https://mvnrepository.com/artifact/com.aliyun/imageseg20191230 -->
<dependency>
      <groupId>com.aliyun</groupId>
      <artifactId>imageseg20191230</artifactId>
      <version>${aliyun.imageseg.version}</version>
</dependency>
*/

import com.aliyun.imageseg20191230.models.SegmentBodyResponse;
import com.aliyun.tea.TeaException;
import com.aliyun.tea.TeaModel;
import java.io.InputStream;
import java.net.URL;

public class SegmentBody {
    public static com.aliyun.imageseg20191230.Client createClient(String accessKeyId, String accessKeySecret) throws Exception {
        /*
          初始化配置对象com.aliyun.teaopenapi.models.Config
          Config对象存放 AccessKeyId、AccessKeySecret、endpoint等配置
         */
         com.aliyun.teaopenapi.models.Config config = new com.aliyun.teaopenapi.models.Config()
                .setAccessKeyId(accessKeyId)
                .setAccessKeySecret(accessKeySecret);
        // 访问的域名
        config.endpoint = "imageseg.cn-shanghai.aliyuncs.com";
        return new com.aliyun.imageseg20191230.Client(config);
    }

    public static void main(String[] args_) throws Exception {
        // 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html
        // 如果您使用的是RAM用户的AccessKey，还需要为子账号授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html
        // 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行代码示例前必须先配置环境变量。
        String accessKeyId = System.getenv("ALIBABA_CLOUD_ACCESS_KEY_ID");
        String accessKeySecret = System.getenv("ALIBABA_CLOUD_ACCESS_KEY_SECRET"); 
        com.aliyun.imageseg20191230.Client client = SegmentBody.createClient(accessKeyId, accessKeySecret);
        // 场景一，使用本地文件
        // InputStream inputStream = new FileInputStream(new File("/tmp/detectVideoShot.mp4"));
        // 场景二，使用任意可访问的url
        URL url = new URL("https://viapi-test-bj.oss-cn-beijing.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody9.png");
        InputStream inputStream = url.openConnection().getInputStream();
        com.aliyun.imageseg20191230.models.SegmentBodyAdvanceRequest segmentBodyAdvanceRequest = new com.aliyun.imageseg20191230.models.SegmentBodyAdvanceRequest()
                .setImageURLObject(inputStream);
        com.aliyun.teautil.models.RuntimeOptions runtime = new com.aliyun.teautil.models.RuntimeOptions();
        try {
            SegmentBodyResponse segmentBodyResponse = client.segmentBodyAdvance(segmentBodyAdvanceRequest, runtime);
            // 获取整体结果
            System.out.println(com.aliyun.teautil.Common.toJSONString(TeaModel.buildMap(segmentBodyResponse)));
            // 获取单个字段
            System.out.println(segmentBodyResponse.getBody().getData().getImageURL());
        } catch (TeaException teaException) {
            // 获取整体报错信息
            System.out.println(com.aliyun.teautil.Common.toJSONString(teaException));
            // 获取单个字段
            System.out.println(teaException.getCode());
        }
    }
}

Python

# -*- coding: utf-8 -*-
# 引入依赖包
# pip install alibabacloud_imageseg20191230

import os
import io
from urllib.request import urlopen
from alibabacloud_imageseg20191230.client import Client
from alibabacloud_imageseg20191230.models import SegmentBodyAdvanceRequest
from alibabacloud_tea_openapi.models import Config
from alibabacloud_tea_util.models import RuntimeOptions

# 初始化Config
config = Config(
    # 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html。
    # 如果您用的是RAM用户的AccessKey，还需要为RAM用户授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html。
    # 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行代码示例前必须先配置环境变量。
    access_key_id=os.environ.get('ALIBABA_CLOUD_ACCESS_KEY_ID'),
    access_key_secret=os.environ.get('ALIBABA_CLOUD_ACCESS_KEY_SECRET'),
    # 访问的域名。
    endpoint='imageseg.cn-shanghai.aliyuncs.com',
    # 访问的域名对应的region
    region_id='cn-shanghai'
)
request = SegmentBodyAdvanceRequest()
#场景一：文件在本地
#stream = open(r'/tmp/SegmentBody9.png', 'rb')
#request.image_urlobject = stream

#场景二：使用任意可访问的url
url = 'https://viapi-test-bj.oss-cn-beijing.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody9.png'
img = urlopen(url).read()
request.image_urlobject = io.BytesIO(img)
request.return_form = 'mask'

runtime_option = RuntimeOptions()
try:
  # 初始化Client
  client = Client(config)
  response = client.segment_body_advance(request, runtime_option)
  # 获取整体结果
  print(response.body)
except Exception as error:
  # 获取整体报错信息
  print(error)
  # 获取单个字段
  print(error.code)
  # tips: 可通过error.__dict__查看属性名称

#关闭流
#stream.close()

PHP

<?php

//安装依赖包
//composer require alibabacloud/imageseg-20191230

use AlibabaCloud\SDK\Imageseg\V20191230\Imageseg;
use AlibabaCloud\Tea\Utils\Utils;
use \Exception;
use Darabonba\OpenApi\Models\Config;
use AlibabaCloud\SDK\Imageseg\V20191230\Models\SegmentBodyAdvanceRequest;
use AlibabaCloud\Tea\Utils\Utils\RuntimeOptions;
use GuzzleHttp\Psr7\Stream;

class SegmentBodyAdvance {

    /**
     * 使用AK&SK初始化账号Client
     * @param string $accessKeyId
     * @param string $accessKeySecret
     * @return Imageseg Client
     */
    public static function createClient($accessKeyId, $accessKeySecret){
        //初始化配置对象Darabonba\OpenApi\Models\Config。
        //Config对象存放accessKeyId、accessKeySecret、endpoint等配置
        $config = new Config([
            "accessKeyId" => $accessKeyId,
            "accessKeySecret" => $accessKeySecret
        ]);
        // 访问的域名
        $config->endpoint = "imageseg.cn-shanghai.aliyuncs.com";
        return new Imageseg($config);
    }
    /**
     * @param string[] $args
     * @return void
     */
    public static function main($args){
        //创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html
        //如果您是用的子账号AccessKey，还需要为子账号授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html
        // 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行代码示例前必须先配置环境变量。
        $accessKeyId = getenv('ALIBABA_CLOUD_ACCESS_KEY_ID');
        $accessKeySecret = getenv('ALIBABA_CLOUD_ACCESS_KEY_SECRET'); 
        $client = self::createClient($accessKeyId, $accessKeySecret);
         // 场景一，使用本地文件
        //$file = fopen('/tmp/SegmentBody9.png', 'rb');
        //$stream = new Stream($file);
        // 场景二，使用任意可访问的url
        $file = fopen('https://viapi-test-bj.oss-cn-beijing.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody9.png', 'rb');
        $stream = new Stream($file);
        $segmentBodyAdvanceRequest = new SegmentBodyAdvanceRequest([
             "imageURLObject" => $stream,
             "returnForm" => "mask"
        ]);
        $runtime = new RuntimeOptions([]);
        try {
            $resp = $client->segmentBodyAdvance($segmentBodyAdvanceRequest, $runtime);
            # 获取整体结果
            echo Utils::toJSONString($resp->body);
        } catch (Exception $exception) {
            # 获取整体报错信息
            echo Utils::toJSONString($exception);
            # 获取单个字段
            echo $exception->getCode();
        }
    }
}
$path = __DIR__ . \DIRECTORY_SEPARATOR . '..' . \DIRECTORY_SEPARATOR . 'vendor' . \DIRECTORY_SEPARATOR . 'autoload.php';
if (file_exists($path)) {
    require_once $path;
}
//$argv是预留的数组入参参数，无实际意义，无需进行修改
SegmentBodyAdvance::main(array_slice($argv, 1));

Node.js

// 安装依赖包
// npm install @alicloud/imageseg20191230
const ImagesegClient = require('@alicloud/imageseg20191230');
const OpenapiClient = require('@alicloud/openapi-client');
const TeaUtil = require('@alicloud/tea-util');
const fs = require('fs');
const http = require('http');
const https = require('https');

let config = new OpenapiClient.Config({
    // 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html。
    // 如果您用的是RAM用户AccessKey，还需要为RAM用户授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html。
    // 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行示例前必须先配置环境变量。 
    accessKeyId: process.env.ALIBABA_CLOUD_ACCESS_KEY_ID,   
    accessKeySecret: process.env.ALIBABA_CLOUD_ACCESS_KEY_SECRET
});
// 访问的域名
config.endpoint = `imageseg.cn-shanghai.aliyuncs.com`;
const client = new ImagesegClient.default(config);
const getResponse = function (httpClient, url) {
  return new Promise((resolve, reject) => {
    httpClient.get(url, function (response) {
      resolve(response);
    })
  })
}
const request = async function () {
  try {
    let segmentBodyAdvanceRequest = new ImagesegClient.SegmentBodyAdvanceRequest();
    // 场景一，使用本地文件
    // const fileStream = fs.createReadStream('/tmp/SegmentBody1.png');
    // segmentBodyAdvanceRequest.imageURLObject = fileStream;
    // 场景二，使用任意可访问的url
    const url = new URL("https://viapi-test-bj.oss-cn-beijing.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody9.png");
    const httpClient = (url.protocol == "https:") ? https : http; 
    const request = httpClient.get(url, function(response) {
    segmentBodyAdvanceRequest.imageURLObject = await getResponse(httpClient, url);
    let runtime = new TeaUtil.RuntimeOptions({ });
    client.segmentBodyAdvance(segmentBodyAdvanceRequest, runtime)
      .then(function(segmentBodyResponse) {
        // 获取整体结果
        console.log(segmentBodyResponse);
        // 获取单个字段
        console.log(segmentBodyResponse.body.data);
      }, function(error) {
        // 获取整体报错信息
        console.log(error);
        // 获取单个字段
        console.log(error.data.Code);
      })
  } catch (error) {
    console.log(error);
  }
}();

/**
依赖github.com/alibabacloud-go/imageseg-20191230/v2
建议使用go mod tidy安装依赖
 */

import (
	"fmt"
	openapi "github.com/alibabacloud-go/darabonba-openapi/v2/client"
	imageseg20191230 "github.com/alibabacloud-go/imageseg-20191230/v2/client"
	util "github.com/alibabacloud-go/tea-utils/v2/service"
	"github.com/alibabacloud-go/tea/tea"
	"net/http"
)

func main() {
  // 从环境变量读取配置的AccessKey ID
  // 运行示例前必须先配置环境变量  
  // 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html。
  // 如果您用的是RAM用户的AccessKey，还需要为RAM用户授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html。
  // 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行示例前必须先配置环境变量。  
  accessKeyId := os.Getenv("ALIBABA_CLOUD_ACCESS_KEY_ID")  
  accessKeySecret := os.Getenv("ALIBABA_CLOUD_ACCESS_KEY_SECRET")
  // 初始化配置对象 &openapi.Config。Config对象存放AccessKeyId、AccessKeySecret、Endpoint等配置。
  config := &openapi.Config{
    AccessKeyId: tea.String(accessKeyId),
    AccessKeySecret: tea.String(accessKeySecret)
	}
	// 访问的域名
	config.Endpoint = tea.String("imageseg.cn-shanghai.aliyuncs.com")
	client, err := imageseg20191230.NewClient(config)
	if err != nil {
		panic(err)
	}
	// 场景一，使用本地文件
	//file, err := os.Open("/tmp/SegmentBody.png")
	//if err != nil {
	//	fmt.Println("can not open file", err)
	//	panic(err)
	//}
	//segmentBodyAdvanceRequest := &imageseg20191230.SegmentBodyAdvanceRequest{
	//	ImageURLObject: file,
	//}
	// 场景二，使用任意可访问的url
	httpClient := http.Client{}
	file, _ := httpClient.Get("https://viapi-test-bj.oss-cn-beijing.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody9.png")
	segmentBodyAdvanceRequest := &imageseg20191230.SegmentBodyAdvanceRequest{
		ImageURLObject: file.Body,
	}
	runtime := &util.RuntimeOptions{}
	segmentBodyAdvanceResponse, err := client.SegmentBodyAdvance(segmentBodyAdvanceRequest, runtime)
	if err != nil {
		// 获取整体报错信息
		fmt.Println(err.Error())
	} else {
		// 获取整体结果
		fmt.Println(segmentBodyAdvanceResponse)
	}
}

// 安装依赖包
// dotnet add package AlibabaCloud.SDK.Imageseg20191230
using System;
using System.Collections;
using System.Collections.Generic;
using System.IO;
using System.Threading.Tasks;
using AlibabaCloud.SDK.Imageseg20191230.Models;
using Tea;
using Tea.Utils;

namespace AlibabaCloud.SDK.Sample
{
    public class Sample
    {
        /**
        * 使用AK&SK初始化账号Client
        * @param accessKeyId
        * @param accessKeySecret
        * @return Client
        * @throws Exception
        */
        public static AlibabaCloud.SDK.Imageseg20191230.Client CreateClient(string accessKeyId, string accessKeySecret)
        {
            AlibabaCloud.OpenApiClient.Models.Config config = new AlibabaCloud.OpenApiClient.Models.Config
            {
              AccessKeyId = accessKeyId,
              AccessKeySecret = accessKeySecret,
            };
            // 访问的域名
            config.Endpoint = "imageseg.cn-shanghai.aliyuncs.com";
            return new AlibabaCloud.SDK.Imageseg20191230.Client(config);
        }
        public static void Main(string[] args)
        {
            // 创建AccessKey ID和AccessKey Secret，请参考https://help.aliyun.com/document_detail/175144.html
            // 如果您使用的是RAM用户的AccessKey，还需要为子账号授予权限AliyunVIAPIFullAccess，请参考https://help.aliyun.com/document_detail/145025.html
            // 从环境变量读取配置的AccessKey ID和AccessKey Secret。运行代码示例前必须先配置环境变量。
            AlibabaCloud.SDK.Imageseg20191230.Client client = CreateClient(System.Environment.GetEnvironmentVariable("ALIBABA_CLOUD_ACCESS_KEY_ID"), System.Environment.GetEnvironmentVariable("ALIBABA_CLOUD_ACCESS_KEY_SECRET"));
            AlibabaCloud.SDK.Imageseg20191230.Models.SegmentBodyAdvanceRequest segmentBodyAdvanceRequest = new AlibabaCloud.SDK.Imageseg20191230.Models.SegmentBodyAdvanceRequest
            ();
            // 场景一，使用本地文件
            System.IO.StreamReader file = new System.IO.StreamReader(@"/tmp/SegmentBody1.png");
            segmentBodyAdvanceRequest.ImageURLObject = file.BaseStream;

            // 场景二，使用任意可访问的url
            //HttpWebRequest request = (HttpWebRequest)WebRequest.Create("https://viapi-test-bj.oss-cn-beijing.aliyuncs.com/viapi-3.0domepic/imageseg/SegmentBody/SegmentBody9.png");
            //WebResponse response = request.GetResponse();
            //Stream stream = response.GetResponseStream();
            //segmentBodyAdvanceRequest.ImageURLObject = stream;
            AlibabaCloud.TeaUtil.Models.RuntimeOptions runtime = new AlibabaCloud.TeaUtil.Models.RuntimeOptions();
            try
            {
                AlibabaCloud.SDK.Imageseg20191230.Models.SegmentBodyResponse SegmentBodyResponse = client.SegmentBodyAdvance(segmentBodyAdvanceRequest, runtime);
                // 获取整体结果
                Console.WriteLine(AlibabaCloud.TeaUtil.Common.ToJSONString(SegmentBodyResponse.Body));
                // 获取单个字段
                Console.WriteLine(AlibabaCloud.TeaUtil.Common.ToJSONString(SegmentBodyResponse.Body.Data));
            }
            catch (TeaException error)
            {
                // 如有需要，请打印 error
                Console.WriteLine(AlibabaCloud.TeaUtil.Common.AssertAsString(error.Message));
            }
            catch (Exception _error)
            {
                TeaException error = new TeaException(new Dictionary<string, object>
              {
                { "message", _error.Message }
              });
                // 如有需要，请打印 error
                Console.WriteLine(AlibabaCloud.TeaUtil.Common.AssertAsString(error.Message));
            }
        }
    }
}

常见的处理后结果

证件照制作

调用人体分割之后可以拿到人体与背景分割后的前景人体图（4通道），可以结合人脸检测与五官定位与红、白、蓝背景融合制作证件照。具体操作步骤如下：

调用人脸检测与五官定位SDK获取人脸105关键点信息。
通过分割SDK获取分割png图。
根据人脸关键点进行人脸对齐，返回旋转后图像、面部中心点、旋转角度、旋转后新的landmarks值。
输入旋转后图片、预期的证件照分辨率、landmarks、背景色，返回半身像结果。
证件照背景色替换。

此处以python为例进行说明。

安装相关依赖

pip install alibabacloud_facebody20191230
pip install opencv-python
pip install oss2
pip install aliyun-python-sdk-viapiutils
pip install viapi-utils

完整代码如下

import argparse
import math
import cv2

from typing import List

import numpy as np
from alibabacloud_imageseg20191230.client import Client as imageseg20191230Client
from alibabacloud_facebody20191230.client import Client as facebody20191230Client
from alibabacloud_tea_openapi import models as open_api_models
from alibabacloud_facebody20191230 import models as facebody_20191230_models
from alibabacloud_imageseg20191230 import models as imageseg_20191230_models
from alibabacloud_tea_util import models as util_models
from alibabacloud_tea_util.client import Client as UtilClient
from viapi.fileutils import FileUtils

import urllib.request


class Sample:
    def __init__(self):
        pass

    @staticmethod
    def create_client(
            access_key_id: str,
            access_key_secret: str,
    ) -> (imageseg20191230Client, facebody20191230Client):
        """
        使用AK&SK初始化账号Client
        @param access_key_id:
        @param access_key_secret:
        @return: Client
        @throws Exception
        """
        seg_config = open_api_models.Config(
            # 您的 AccessKey ID,
            access_key_id=access_key_id,
            # 您的 AccessKey Secret,
            access_key_secret=access_key_secret
        )
        face_config = open_api_models.Config(
            # 您的 AccessKey ID,
            access_key_id=access_key_id,
            # 您的 AccessKey Secret,
            access_key_secret=access_key_secret
        )
        # 访问的域名
        seg_config.endpoint = f'imageseg.cn-shanghai.aliyuncs.com'
        face_config.endpoint = f'facebody.cn-shanghai.aliyuncs.com'
        return (imageseg20191230Client(seg_config), facebody20191230Client(face_config))

def id_pohto_demo(
        args: List[str],
) -> None:
    assert (args is not None and len(args.param) > 2), "parameters wrong, use -h for details!"
    [sc_file, color, of_file] = [i for i in args.param]

    # set id photo size
    size = [295, 413]

    # client init
    seg_client, face_client = Sample.create_client(args.access_key_id, args.access_key_secret)

    # generation source image url
    input_url = generate_url(sc_file, args.access_key_id, args.access_key_secret)

    # init segment_body_request
    segment_body_request = imageseg_20191230_models.SegmentBodyRequest(
        image_url=input_url
    )

    # init detect_face_request
    detect_face_request = facebody_20191230_models.DetectFaceRequest(
        image_url=input_url
    )
    runtime = util_models.RuntimeOptions()
    try:
        # 通过人脸SDK获取人脸105关键点信息
        face_result = face_client.detect_face_with_options(detect_face_request, runtime)
        landmarks = face_result.body.data.landmarks
        landmarks = np.array(landmarks)

        # 通过分割SDK获取分割png图
        seg_result = seg_client.segment_body_with_options(segment_body_request, runtime)
        print(seg_result.body.data.image_url)
        rqt = urllib.request.urlopen(seg_result.body.data.image_url)
        seg_img = np.asarray(bytearray(rqt.read()), dtype="uint8")
        seg_img = cv2.imdecode(seg_img, cv2.IMREAD_UNCHANGED)

        # 根据人脸关键点进行人脸对齐，返回旋转后图像、面部中心点、旋转角度、旋转后新的landmarks值
        rotated_img, eye_center, angle, landmarks = align_face(seg_img, landmarks)

        # 输入旋转后图片、预期的证件照分辨率、landmarks、背景色，返回半身像结果
        png_img = corp_halfbody(rotated_img, landmarks, size)

        # 证件照背景色替换
        colors = {'red': (0, 0, 255, 255), 'blue': (255, 0, 0, 255), 'white': (255, 255, 255, 255)}
        if type(color) is str:
            color = colors[color]
        rst_img = np.zeros((size[1], size[0], 3)) + color[0:3]
        rst_img = image_merge_background(png_img[:, :, 0:3], png_img, rst_img)

        # 保存结果到本地
        cv2.imwrite(of_file, rst_img)
    except Exception as error:
        print(error.message)
        # 如有需要，请打印 error
        UtilClient.assert_as_string(error.message)

def align_face(image_array, landmarks):
    """ align faces according to eyes position
    :param image_array: numpy array of a single image
    :param landmarks: dict of landmarks for facial parts as keys and tuple of coordinates as values
    :return:
    rotated_img:  numpy array of aligned image
    eye_center: tuple of coordinates for eye center
    angle: degrees of rotation
    """
    landmarks = np.resize(landmarks, (105, 2))

    # get list landmarks of left and right eye
    left_eye = landmarks[24:39]
    right_eye = landmarks[40:55]
    left_eye_center = np.mean(left_eye, axis=0).astype("int32")
    right_eye_center = np.mean(right_eye, axis=0).astype("int32")
    dy = right_eye_center[1] - left_eye_center[1]
    dx = right_eye_center[0] - left_eye_center[0]
    angle = math.atan2(dy, dx) * 180. / math.pi
    # calculate the center of 2 eyes
    eye_center = (int(left_eye_center[0] + right_eye_center[0]) // 2,
                  int(left_eye_center[1] + right_eye_center[1]) // 2)
    # at the eye_center, rotate the image by the angle
    rotate_matrix = cv2.getRotationMatrix2D(eye_center, angle, scale=1)
    rotated_img = cv2.warpAffine(image_array, rotate_matrix, (image_array.shape[1], image_array.shape[0]))

    rotated_landmarks = []
    for landmark in landmarks:
        rotated_landmark = rotate(origin=eye_center, point=landmark, angle=angle, row=image_array.shape[0])
        rotated_landmarks.append(rotated_landmark)
    return rotated_img, eye_center, angle, rotated_landmarks

def rotate(origin, point, angle, row):
    """ rotate coordinates in image coordinate system
    :param origin: tuple of coordinates,the rotation center
    :param point: tuple of coordinates, points to rotate
    :param angle: degrees of rotation
    :param row: row size of the image
    :return: rotated coordinates of point
    """
    x1, y1 = point
    x2, y2 = origin
    y1 = row - y1
    y2 = row - y2
    angle = math.radians(angle)
    x = x2 + math.cos(angle) * (x1 - x2) - math.sin(angle) * (y1 - y2)
    y = y2 + math.sin(angle) * (x1 - x2) + math.cos(angle) * (y1 - y2)
    y = row - y
    return int(x), int(y)

def corp_halfbody(image_array, landmarks, size):
    """ crop face according to eye,mouth and chin position
    :param image_array: numpy array of a single image
    :param size: single int value, size for w and h after crop
    :param landmarks: dict of landmarks for facial parts as keys and tuple of coordinates as values
    :return:
    cropped_img: numpy array of cropped image
    left, top: left and top coordinates of cropping
    """

    crop_size = [0, 0, size[1], size[0]]

    scal = size[1] / 4 / abs(landmarks[98][1] - landmarks[56][1])
    image = cv2.resize(image_array, np.multiply((image_array.shape[0:2])[::-1], scal).astype(int))
    landmarks = np.multiply(np.array(landmarks), scal)

    x_center = landmarks[98][0]
    crop_size[0:2] = [x_center - size[0] / 2, x_center + size[0] / 2]

    y_center = (landmarks[98][1] + landmarks[56][1]) / 2
    crop_size[2:4] = [y_center - size[1] / 2, y_center + size[1] / 2]

    left, right, top, bottom = [round(i) for i in crop_size]
    left = max(0, left)
    top = max(0, top)
    right = min(image.shape[1], right)
    bottom = min(image.shape[0], bottom)

    cropped_img = image[top:bottom, left:right]

    left, right, top, bottom = [round(i) for i in crop_size]
    bottom = size[1]
    top = size[1] - cropped_img.shape[0]
    left = -min(0, left)
    right = min(cropped_img.shape[1] + left, size[0])


    png_img = np.zeros((size[1], size[0], 4))
    png_img[top:bottom, left:right] = cropped_img[:, :]
    return png_img

def image_merge_background(sc_image, png_image, bg_image):
    """ merge foreground to background image
    :param sc_image: numpy array of the source image
    :param png_image: numpy array of the segmented result with same size of sc_image
    :param bg_image: numpy array of the background image with same size of sc_image
    :return:
    rst_image: numpy array of merged image
    """
    assert (sc_image is not None and png_image is not None and bg_image is not None), "read image input error!"
    h, w, c = sc_image.shape

    # keep sc_image, png_image and bg_image same size
    viapi_image = cv2.resize(png_image, (w, h))
    bg_image = cv2.resize(bg_image, (w, h))
    if len(viapi_image.shape) == 2:
        mask = viapi_image[:, :, np.newaxis]
    elif viapi_image.shape[2] == 4:
        mask = viapi_image[:, :, 3:4]
    elif viapi_image.shape[2] == 3:
        mask = viapi_image[:, :, 0:1]
    else:
        raise Exception("invalid image mask!")
    mask = mask / 255.0

    # merge background
    sc_image = sc_image.astype(float)
    bg_image = bg_image.astype(float)
    rst_image = (sc_image - bg_image) * mask + bg_image
    rst_image = np.clip(rst_image, 0, 255)
    return rst_image

# see https://help.aliyun.com/document_detail/155645.html for more detail
def generate_url(image_path, access_key_id, access_key_secret):
    file_utils = FileUtils(access_key_id, access_key_secret)
    oss_url = file_utils.get_oss_url(image_path, "jpeg", True)
    return oss_url

def define_options():
    parser = argparse.ArgumentParser(description='Human segmentation examples.')
    parser.add_argument('-i', '--access_key_id', default="", help="oss access key id")
    parser.add_argument('-s', '--access_key_secret', default="", help="access key secret")
    parser.add_argument('-p', '--param', nargs=3
                            , default=['./data/source.jpeg', 'red', 'result.jpg']
                            , type=str
                            , help=('source_image_path background_color[red blue white] result_path'))
    args = parser.parse_args()
    return args

if __name__ == '__main__':
    args = define_options()
    id_pohto_demo(args)

原图：

与红背景融合后的证件照图：

证件照的相关代码，请参见idcard。

说明

代码解压后运行python idcard.py -i YOUR_ACCESS_KEY_ID -s YOUR_ACCESS_KEY_SECRET即可，其中YOUR_ACCESS_KEY_ID、YOUR_ACCESS_KEY_SECRET生成操作，请参见创建AccessKey。如果您使用的是子账号AccessKey，您需要给子账号赋予AliyunVIAPIFullAccess权限，具体操作，请参见使用RAM授权。
您的账号还需要开通分割抠图服务和人脸人体服务，若未开通服务请立即开通分割抠图服务和立即开通人脸人体服务。