IPC content understanding

Prerequisites

1. You have purchased the service by creating an IPC order.

2. You have set up the environment. For details, see OpenAPI portal tool.

Basic usage

Create an IPC algorithm template

The following example shows how to call the CreateCustomTemplate operation to create a custom template:

# -*- coding: utf-8 -*-
from alibabacloud_ice20201109.client import Client as ICE20201109Client
from alibabacloud_credentials.client import Client as CredentialClient
from alibabacloud_tea_openapi import models as open_api_models
from alibabacloud_ice20201109 import models as ice20201109_models
from alibabacloud_tea_util import models as util_models

# For production environments, we recommend using a more secure method for authentication that does not involve access keys (AKs). For more information about how to configure credentials, see https://help.aliyun.com/document_detail/378659.html.
credential = CredentialClient()
config = open_api_models.Config(
    credential=credential
)
# For more information about endpoints, see https://api.aliyun.com/product/ICE
config.endpoint = 'ice.cn-shanghai.aliyuncs.com'
client = ICE20201109Client(config)

create_custom_template_request = ice20201109_models.CreateCustomTemplateRequest(
    name='ipc_template_001',
    template_config='{"AnalyseTypes":"vlEventDesc","Industry":"common","Scene":"search","TemplateName":"ipc_template_001"}',
    type=11
)
runtime = util_models.RuntimeOptions()
response = client.create_custom_template_with_options(create_custom_template_request, runtime)
print(response)

Request parameters:

Response parameters:

Response example:

{
  "CustomTemplate": {
    "CreateTime": "2025-12-22T03:11:02Z",
    "IsDefault": false,
    "ModifiedTime": "2025-12-22T03:11:02Z",
    "Status": "Normal",
    "TemplateConfig": "{\"AnalyseTypes\":\"vlEventDesc\"}",
    "TemplateId": "****a0da83434f0a8d57ffb0415c****",
    "TemplateName": "ipc_template_001",
    "Type": 11,
    "TypeName": "AISmarttagTemplate"
  },
  "RequestId": "******22-1193-5A70-8D4C-E0533C******"
}

Submit a smart tag job

The following example shows how to call the SubmitSmarttagJob operation to submit a smart tag job:

# -*- coding: utf-8 -*-
from alibabacloud_ice20201109.client import Client as ICE20201109Client
from alibabacloud_credentials.client import Client as CredentialClient
from alibabacloud_tea_openapi import models as open_api_models
from alibabacloud_ice20201109 import models as ice20201109_models
from alibabacloud_tea_util import models as util_models
import json

# For production environments, we recommend using a more secure method for authentication that does not involve access keys (AKs). For more information about how to configure credentials, see https://help.aliyun.com/document_detail/378659.html.
credential = CredentialClient()
config = open_api_models.Config(
    credential=credential
)
# For more information about endpoints, see https://api.aliyun.com/product/ICE
config.endpoint = 'ice.cn-shanghai.aliyuncs.com'
client = ICE20201109Client(config)

submit_smarttag_job_request_input = ice20201109_models.SubmitSmarttagJobRequestInput(
    type='Multi',
    media=json.dumps(
        {
            "type": "URL",
            "inputs": [
                "https://xxx.png",
                "https://xxx.png"
            ]
        }
    )
)
submit_smarttag_job_request = ice20201109_models.SubmitSmarttagJobRequest(
    title='ipc-test-001',
    input=submit_smarttag_job_request_input,
    template_id='xxx',
    # params='{"vlEventDescParams":{"deviceId":"camera_001"}}',
)
runtime = util_models.RuntimeOptions()
response = client.submit_smarttag_job_with_options(submit_smarttag_job_request, runtime)
print(response)

Request parameters:

Description of the vlEventDescParams field:

Valid values for targetLanguage:

Response parameters:

Response example:

{
    "RequestId": "******11-DB8D-4A9A-875B-275798******",
    "JobId": "****d80e4e4044975745c14b****"
}

Query smart tag job results

The following example shows how to call the QuerySmarttagJob operation to query job results:

# -*- coding: utf-8 -*-
from alibabacloud_ice20201109.client import Client as ICE20201109Client
from alibabacloud_credentials.client import Client as CredentialClient
from alibabacloud_tea_openapi import models as open_api_models
from alibabacloud_ice20201109 import models as ice20201109_models
from alibabacloud_tea_util import models as util_models

# For production environments, we recommend using a more secure method for authentication that does not involve access keys (AKs). For more information about how to configure credentials, see https://help.aliyun.com/document_detail/378659.html.
credential = CredentialClient()
config = open_api_models.Config(
    credential=credential
)
# For more information about endpoints, see https://api.aliyun.com/product/ICE
config.endpoint = 'ice.cn-shanghai.aliyuncs.com'
client = ICE20201109Client(config)

query_smarttag_job_request = ice20201109_models.QuerySmarttagJobRequest(
    job_id='xxx'
)
runtime = util_models.RuntimeOptions()
response = client.query_smarttag_job_with_options(query_smarttag_job_request, runtime)
print(response)

Request parameters:

Response parameters:

Data fields:

Response example:

{
  "JobStatus": "Success",
  "RequestId": "41B38CF1-C1EC-538E-81FB-791355065538",
  "Results": {
    "Result": [
      {
        "Data": "\"{\\\"brief\\\":\\\"An adult male is sitting on a chair eating, while two dogs in a nearby cage are moving around.\\\",\\\"caption\\\":\\\"In a brick-paved courtyard, an adult male (short hair, wearing a beige robe) is sitting on a chair and eating. Next to him is a table with food and a water bottle. Two medium-sized, grayish-white dogs are moving around in a cage.\\\",\\\"action\\\":[\\\"eating\\\",\\\"moving\\\"],\\\"label\\\":{\\\"package\\\":false,\\\"human\\\":true,\\\"pet\\\":true,\\\"vehicle\\\":false},\\\"title\\\":\\\"Man eats in courtyard, dogs move in cage\\\",\\\"event\\\":[\\\"adult male eating\\\",\\\"dog moving\\\"],\\\"object\\\":[\\\"adult male\\\",\\\"dog\\\",\\\"table\\\",\\\"chair\\\"]}\"",
        "Type": "EventDesc"
      }
    ]
  },
  "UserData": "{\"fromICE\":true,\"pk\":1346693276669150}"
}

Advanced usage

Advanced features let you identify specific pedestrians and pets.

Create a custom recognition library

Call the CreateCustomRecognitionLib operation to create a custom recognition library. The following code provides an example:

# -*- coding: utf-8 -*-
from alibabacloud_ice20201109.client import Client as ICE20201109Client
from alibabacloud_credentials.client import Client as CredentialClient
from alibabacloud_tea_openapi import models as open_api_models
from alibabacloud_ice20201109 import models as ice20201109_models
from alibabacloud_tea_util import models as util_models

# For production environments, we recommend that you use a more secure method that does not involve access keys (AKs) for authentication. For more information about how to configure credentials, see https://help.aliyun.com/document_detail/378659.html.
credential = CredentialClient()
config = open_api_models.Config(
    credential=credential
)
# For more information about endpoints, see https://api.aliyun.com/product/ICE.
config.endpoint = 'ice.cn-shanghai.aliyuncs.com'
client = ICE20201109Client(config)

create_recognition_lib_request = ice20201109_models.CreateRecognitionLibRequest(
    algorithm='pedestrian',
    lib_name='ipc_peds_lib_001'
)
runtime = util_models.RuntimeOptions()
response = client.create_recognition_lib_with_options(create_recognition_lib_request, runtime)
print(response)

Request parameters:

Response parameters:

Response example:

{
    "LibId": "*************24b47865c6**************",
    "RequestId": "******11-DB8D-4A9A-875B-275798******"
}

Create a custom recognition entity

Call the CreateRecognitionEntity operation to create an entity in the library. The following code provides an example:

# -*- coding: utf-8 -*-
from alibabacloud_ice20201109.client import Client as ICE20201109Client
from alibabacloud_credentials.client import Client as CredentialClient
from alibabacloud_tea_openapi import models as open_api_models
from alibabacloud_ice20201109 import models as ice20201109_models
from alibabacloud_tea_util import models as util_models

# For production environments, we recommend that you use a more secure method that does not involve access keys (AKs) for authentication. For more information about how to configure credentials, see https://help.aliyun.com/document_detail/378659.html.
credential = CredentialClient()
config = open_api_models.Config(
    credential=credential
)
# For more information about endpoints, see https://api.aliyun.com/product/ICE.
config.endpoint = 'ice.cn-shanghai.aliyuncs.com'
client = ICE20201109Client(config)

create_recognition_entity_request = ice20201109_models.CreateRecognitionEntityRequest(
    algorithm='pedestrian',
    lib_id='xxx',
    entity_name='Xiaoshuai'
)
runtime = util_models.RuntimeOptions()
response = client.create_recognition_entity_with_options(create_recognition_entity_request, runtime)
print(response)

Request parameters:

Response parameters:

Response example:

{
  "EntityId": "**************544cb84754************",
  "RequestId": "******11-DB8D-4A9A-875B-275798******"
}

Create a custom recognition sample

Call the CreateRecognitionSample operation to create a sample for an entity. The following code provides an example:

# -*- coding: utf-8 -*-

from alibabacloud_ice20201109.client import Client as ICE20201109Client
from alibabacloud_credentials.client import Client as CredentialClient
from alibabacloud_tea_openapi import models as open_api_models
from alibabacloud_ice20201109 import models as ice20201109_models
from alibabacloud_tea_util import models as util_models

# For production environments, we recommend that you use a more secure method that does not involve access keys (AKs) for authentication. For more information about how to configure credentials, see https://help.aliyun.com/document_detail/378659.html.
credential = CredentialClient()
config = open_api_models.Config(
    credential=credential
)
# For more information about endpoints, see https://api.aliyun.com/product/ICE.
config.endpoint = 'ice.cn-shanghai.aliyuncs.com'
client = ICE20201109Client(config)

create_recognition_sample_request = ice20201109_models.CreateRecognitionSampleRequest(
    algorithm='pedestrian',
    lib_id='xxx',
    entity_id='xxx',
    image_url='https://xxx.png'
)
runtime = util_models.RuntimeOptions()
response = client.create_recognition_sample_with_options(create_recognition_sample_request, runtime)
print(response)

Request parameters:

Response parameters:

Response example:

{
    "RequestId": "******11-DB8D-4A9A-875B-275798******",
    "SampleId": "**************4d2ba728e2f**************"
}

Create an IPC algorithm template

Call the CreateCustomTemplate operation to create a template. The process is the same as in the basic usage scenario, but you must also configure a pedestrian library or pet library. The following code provides an example:

# -*- coding: utf-8 -*-
from alibabacloud_ice20201109.client import Client as ICE20201109Client
from alibabacloud_credentials.client import Client as CredentialClient
from alibabacloud_tea_openapi import models as open_api_models
from alibabacloud_ice20201109 import models as ice20201109_models
from alibabacloud_tea_util import models as util_models

# For production environments, we recommend that you use a more secure method that does not involve access keys (AKs) for authentication. For more information about how to configure credentials, see https://help.aliyun.com/document_detail/378659.html.
credential = CredentialClient()
config = open_api_models.Config(
    credential=credential
)
# For more information about endpoints, see https://api.aliyun.com/product/ICE.
config.endpoint = 'ice.cn-shanghai.aliyuncs.com'
client = ICE20201109Client(config)

create_custom_template_request = ice20201109_models.CreateCustomTemplateRequest(
    name='ipc_template_001_with_group',
    template_config='{"AnalyseTypes":"vlEventDesc","Industry":"common","Scene":"search","TemplateName":"ipc_template_001_with_group","PedestrianGroupIds":"xxx"}',
    type=11
)
runtime = util_models.RuntimeOptions()
response = client.create_custom_template_with_options(create_custom_template_request, runtime)
print(response)

Request parameters:

Submit an IPC intelligent tag job

Call the SubmitIntelligentTagJob operation to submit a job. The process is the same as in the basic usage scenario.

Query IPC job results

Call the QueryIntelligentTagJob operation to query the job results. The process is the same as in the basic usage scenario. As the following example shows, the output replaces the generic term pedestrian with the custom name Xiaoshuai:

{
  "JobStatus": "Success",
  "RequestId": "******F9-82A8-5BC4-B63E-63D756******",
  "Results": {
    "Result": [
      {
        "Data": "\"{\\\"brief\\\":\\\"Xiaoshuai is sitting on a chair eating, with two dogs in a nearby cage.\\\",\\\"caption\\\":\\\"In a brick-paved courtyard, an adult male (Xiaoshuai) is sitting on a chair and eating. Next to him is a table with food and a water bottle. Two medium-sized, grayish-white dogs are in a cage.\\\",\\\"action\\\":[\\\"eating\\\",\\\"other events\\\"],\\\"label\\\":{\\\"package\\\":false,\\\"human\\\":true,\\\"pet\\\":true,\\\"vehicle\\\":false},\\\"title\\\":\\\"Xiaoshuai eats in the courtyard while dogs are in the cage\\\",\\\"event\\\":[\\\"adult male eating\\\",\\\"dog other events\\\"],\\\"object\\\":[\\\"adult male\\\",\\\"dog\\\",\\\"table\\\",\\\"chair\\\"]}\"",
        "Type": "EventDesc"
      }
    ]
  },
  "UserData": "{\"fromICE\":true,\"pk\":1346693276669150}"
}