Merge pull request #260341 from MicrosoftDocs/main

12/5/2023 AM Publish

Author: Taojunshen

articles/ai-services/computer-vision/concept-model-customization.md

@@ -23,6 +23,12 @@ Model customization lets you train a specialized Image Analysis model for your o
 > [Python SDK quickstart](./how-to/model-customization.md?tabs=python)
 
 
+> [!IMPORTANT]
+> You can train a custom model using either the [Custom Vision service](/azure/ai-services/custom-vision-service/overview) or the Image Analysis 4.0 service with model customization. The following table compares the two services.
+>
+> [!INCLUDE [custom-vision-ia-compare](includes/custom-vision-ia-compare.md)]
+
+
 ## Scenario components
 
 The main components of a model customization system are the training images, COCO file, dataset object, and model object.

articles/ai-services/computer-vision/concept-shelf-analysis.md

@@ -27,6 +27,12 @@ Try out the capabilities of Product Recognition quickly and easily in your brows
 > [!NOTE]
 > The brands shown in the images are not affiliated with Microsoft and do not indicate any form of endorsement of Microsoft or Microsoft products by the brand owners, or an endorsement of the brand owners or their products by Microsoft.
 
+> [!IMPORTANT]
+> You can train a custom model for product recognition using either the [Custom Vision service](/azure/ai-services/custom-vision-service/overview) or the Image Analysis 4.0 Product Recognition APIs. The following table compares the two services.
+>
+> [!INCLUDE [custom-vision-shelf-compare](includes/custom-vision-shelf-compare.md)]
+
+
 ## Product Recognition features
 
 ### Shelf Image Composition

articles/ai-services/computer-vision/how-to/mitigate-latency.md

@@ -107,7 +107,7 @@ The following are other tips to ensure the reliability and high support of your
 
 ## Next steps
 
-In this guide, you learned how to improve performance when using the Face service. Next, Follow the tutorial to set up a working software solution that combines server-side and client-side logic to do face liveness detection on users.
+In this guide, you learned how to improve performance when using the Face service. Next, follow the tutorial to set up a working software solution that combines server-side and client-side logic to do face liveness detection on users.
 
 > [!div class="nextstepaction"]
 > [Tutorial: Detect face liveness](../Tutorials/liveness.md)

articles/ai-services/computer-vision/identity-quotas-limits.md

@@ -27,6 +27,8 @@ This article contains a reference and a detailed description of the quotas and l
 | Free (F0) | 20 transactions per minute |
 | Standard (S0),</br>Enterprise (E0) | 10 transactions per second, and 200 TPS across all resources in a single region.</br>See the next section if you want to increase this limit. |
 
+> [!NOTE]
+> If you exceed the default rate limit, you'll receive a `429` error. To address this issue, refer to the [Performance guide](/azure/ai-services/computer-vision/how-to/mitigate-latency#handle-errors-effectively).
 
 **Default Face resource quantity limits**
 
@@ -41,7 +43,8 @@ This article contains a reference and a detailed description of the quotas and l
 
 To increase rate limits and resource limits, you can submit a support request. However, for other quota limits, you need to switch to a higher pricing tier to increase those quotas. 
 
-[Submit a support request](/azure/ai-services/cognitive-services-support-options?context=%2Fazure%2Fai-services%2Fopenai%2Fcontext%2Fcontext) and answer the following questions: 
+[Submit a support request](/azure/ai-services/cognitive-services-support-options?context=%2Fazure%2Fai-services%2Fopenai%2Fcontext%2Fcontext) and provide the following information: 
+- A description of your Face use case.
 - The reason for requesting an increase in your current limits. 
 - Which of your subscriptions or resources are affected? 
 - What limits would you like to increase? (rate limits or resource limits) 

articles/ai-services/computer-vision/includes/custom-vision-ia-compare.md

@@ -0,0 +1,24 @@
+---
+title: "Custom Vision Image Analysis comparison table"
+titleSuffix: "Azure AI services"
+services: cognitive-services
+author: PatrickFarley
+manager: nitinme
+
+ms.service: azure-ai-vision
+ms.topic: include
+ms.date: 12/04/2023
+ms.author: pafarley
+---
+
+| Areas | Custom Vision service | Image Analysis 4.0 service |
+| --- | --- | --- |
+|Tasks |Image classification</br>Object detection |Image classification</br>Object detection |
+|Base model |CNN  |Transformer model |
+|Labeling|[Customvision.ai](https://www.customvision.ai/) |AML Studio or COCO file |
+|Web Portal |[Customvision.ai](https://www.customvision.ai/) |[Vision Studio](https://portal.vision.cognitive.azure.com/gallery/featured) |
+|Libraries |REST, SDK |REST, Python Sample |
+|Minimum training data needed |15 images per category |2-5 images per category |
+|Training data storage |Uploaded to service |In customer’s blob storage account |
+|Model hosting |Cloud and Edge |Cloud hosting only, Edge container hosting to come |
+|AI quotient (AIQ) | <table ><col ><col ></colgroup><thead><tr><th>context</th><th>IC</br>(top-1 accuracy, 22 datasets)</th><th>OD</br>(mAP@50, 59 datasets)</th></tr></thead><tbody><tr><td>0 shot</td><td>N/A</td><td>N/A</td></tr><tr><td>2 shot</td><td>51.47</td><td>33.3</td></tr><tr><td>3 shot</td><td>56.73</td><td>37.0</td></tr><tr><td>5 shot</td><td>63.01</td><td>43.4</td></tr><tr><td>10 shot</td><td>68.95</td><td>54.0</td></tr><tr><td>full</td><td>85.25</td><td>76.6</td></tr></tbody></table> | <table ><col ><col ></colgroup><thead><tr><th>context</th><th>IC</br>(top-1 accuracy, 22 datasets)</th><th>OD</br>(mAP@50, 59 datasets)</th></tr></thead><tbody><tr><td>0 shot</td><td>57.8</td><td>27.1</td></tr><tr><td>2 shot</td><td>73.02</td><td>49.2</td></tr><tr><td>3 shot</td><td>75.51</td><td>61.1</td></tr><tr><td>5 shot</td><td>79.14</td><td>68.2</td></tr><tr><td>10 shot</td><td>81.31</td><td>75.0</td></tr><tr><td>full</td><td>90.98</td><td>85.4</td></tr></tbody></table> |

articles/ai-services/computer-vision/includes/custom-vision-shelf-compare.md

@@ -0,0 +1,25 @@
+---
+title: "Custom Vision Shelf Analysis comparison table"
+titleSuffix: "Azure AI services"
+services: cognitive-services
+author: PatrickFarley
+manager: nitinme
+
+ms.service: azure-ai-vision
+ms.topic: include
+ms.date: 12/04/2023
+ms.author: pafarley
+---
+
+|Areas | Products on Shelves – Custom Vision |Product Recognition – Image Analysis API/Customization |
+|---|---|---|
+|Features | Custom product understanding |Image stitching & rectification,</br>Pretrained product understanding,</br>Custom product understanding,</br>Planogram matching |  
+|Base model |CNN |Florence |
+|Labeling |[Customvision.ai](https://www.customvision.ai/) |AML Studio or COCO file |
+|Web Portal |[Customvision.ai](https://www.customvision.ai/) |[Vision Studio](https://portal.vision.cognitive.azure.com/gallery/featured) |
+|Libraries |REST, SDK |REST, Python Sample |
+|Minimum training data needed |15 images per category |2-5 images per category| 
+|Training data storage |Uploaded to service |In customer’s blob storage account |
+|Model hosting |Cloud and Edge |Cloud hosting only, Edge container hosting to come |
+|AI quotient (AIQ)| <table ><colgroup><col ><col ></colgroup><thead><tr><th>context</th><th>Top-1 accuracy, 14 datasets</th></tr></thead><tbody><tr><td>1 shot (catalog)</td><td>29.4</td></tr><tr><td>2 shot</td><td>57.1</td></tr><tr><td>3 shot</td><td>66.7</td></tr><tr><td>5 shot</td><td>80.8</td></tr><tr><td>10 shot</td><td>86.4</td></tr><tr><td>full</td><td>94.9</td></tr></tbody></table>| <table ><colgroup><col ><col ></colgroup><thead><tr><th>context</th><th>Top-1 accuracy, 14 datasets</th></tr></thead><tbody><tr><td>1 shot (catalog)</td><td>86.9</td></tr><tr><td>2 shot</td><td>88.8</td></tr><tr><td>3 shot</td><td>89.8</td></tr><tr><td>5 shot</td><td>90.3</td></tr><tr><td>10 shot</td><td>91.0</td></tr><tr><td>full</td><td>95.4</td></tr></tbody></table>|
+

articles/ai-services/computer-vision/reference-video-search.md

@@ -333,7 +333,6 @@ Represents the create ingestion request model for the JSON document.
 | videos | [ [IngestionDocumentRequestModel](#ingestiondocumentrequestmodel) ] | Gets or sets the list of video document ingestion requests in the JSON document. | No |
 | moderation | boolean | Gets or sets the moderation flag, indicating if the content should be moderated. | No |
 | generateInsightIntervals | boolean | Gets or sets the interval generation flag, indicating if insight intervals should be generated. | No |
-| documentAuthenticationKind | string | Gets or sets the authentication kind that is to be used for downloading the documents.<br>*Enum:* `"none"`, `"managedIdentity"` | No |
 | filterDefectedFrames | boolean | Frame filter flag indicating frames will be evaluated and all defected (e.g. blurry, lowlight, overexposure) frames will be filtered out. | No |
 
 ### DatetimeFilterModel

articles/ai-services/content-moderator/api-reference.md

@@ -14,7 +14,7 @@ ms.author: pafarley
 
 # Content Moderator API reference
 
-You can get started with Azure AI Content Moderator APIs by doing the following:
+You can get started with Azure Content Moderator APIs by doing the following:
 
 - In the Azure portal, [subscribe to the Content Moderator API](https://portal.azure.com/#create/Microsoft.CognitiveServicesContentModerator).
 

articles/ai-services/content-moderator/client-libraries.md

@@ -12,7 +12,7 @@ ms.date: 09/28/2021
 ms.author: pafarley
 ms.devlang: csharp, java, python
 ms.custom: devx-track-python, devx-track-csharp, cog-serv-seo-aug-2020, mode-api, devx-track-dotnet, devx-track-extended-java
-keywords: content moderator, Azure AI Content Moderator, online moderator, content filtering software
+keywords: content moderator, Azure Content Moderator, online moderator, content filtering software
 ---
 
 # Quickstart: Use the Content Moderator client library

articles/ai-services/content-moderator/includes/quickstarts/csharp-sdk.md

@@ -1,7 +1,7 @@
 ---
 title: "Content Moderator .NET client library quickstart"
 titleSuffix: Azure AI services
-description: In this quickstart, learn how to get started with the Azure AI Content Moderator client library for .NET. Build content filtering software into your app to comply with regulations or maintain the intended environment for your users.
+description: In this quickstart, learn how to get started with the Azure Content Moderator client library for .NET. Build content filtering software into your app to comply with regulations or maintain the intended environment for your users.
 #services: cognitive-services
 author: PatrickFarley
 manager: nitinme
@@ -13,7 +13,7 @@ ms.custom: "devx-track-dotnet, cog-serv-seo-aug-2020"
 
 ---
 
-Get started with the Azure AI Content Moderator client library for .NET. Follow these steps to install the NuGet package and try out the example code for basic tasks. 
+Get started with the Azure Content Moderator client library for .NET. Follow these steps to install the NuGet package and try out the example code for basic tasks. 
 
 Content Moderator is an AI service that lets you handle content that is potentially offensive, risky, or otherwise undesirable. Use the AI-powered content moderation service to scan text, image, and videos and apply content flags automatically. Build content filtering software into your app to comply with regulations or maintain the intended environment for your users.