You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Many face recognition issues are caused by low-quality reference images. Some factors that can degrade model performance are:
72
-
* Face size (faces that are distant from the camera)
73
-
* Face orientation (faces turned or tilted away from camera)
74
-
* Poor lighting conditions (either low light or backlighting) where the image may be poorly exposed or have too much noise
75
-
* Occlusion (partially hidden or obstructed faces) including accessories like hats or thick-rimmed glasses)
76
-
* Blur (such as by rapid face movement when the photograph was taken).
71
+
1. Add more instructions to improve verification accuracy.
77
72
78
-
The service provides image quality checks to help you make the choice of whether the image is of sufficient quality based on the above factors to add the customer or attempt face recognition. This app demonstrates how to access frames from the device's camera, detect quality and show user interface messages to the user to help them capture a higher quality image, select the highest-quality frames, and add the detected face into the Face API service.
73
+
Many face recognition issues are caused by low-quality reference images. Some factors that can degrade model performance are:
74
+
* Face size (faces that are distant from the camera)
75
+
* Face orientation (faces turned or tilted away from camera)
76
+
* Poor lighting conditions (either low light or backlighting) where the image may be poorly exposed or have too much noise
77
+
* Occlusion (partially hidden or obstructed faces) including accessories like hats or thick-rimmed glasses)
78
+
* Blur (such as by rapid face movement when the photograph was taken).
79
+
80
+
The service provides image quality checks to help you make the choice of whether the image is of sufficient quality based on the above factors to add the customer or attempt face recognition. This app demonstrates how to access frames from the device's camera, detect quality and show user interface messages to the user to help them capture a higher quality image, select the highest-quality frames, and add the detected face into the Face API service.
Notice the app also offers functionality for deleting the user's information and the option to re-add.
86
+
1. The sample app offers functionality for deleting the user's information and the option to readd. You can enable or disable these operations based on your business requirement.
To extend the app's functionality to cover the full experience, read the [overview](../enrollment-overview.md) for additional features to implement and best practices.
90
92
93
+
1. Configure your database to map each person with their ID
94
+
95
+
You need to use a database to store the face image along with user metadata. The social security number or other unique person identifier can be used as a key to look up their face ID.
96
+
97
+
1. For secure methods of passing your subscription key and endpoint to Face service, see the Azure Cognitive Services [Security](/azure/cognitive-services/cognitive-services-security?tabs=command-line%2Ccsharp) guide.
title: "Overview: Generate alt text of images with Image Analysis"
3
+
titleSuffix: Azure Cognitive Services
4
+
description: Grow your customer base by making your products and services more accessible. Generate a description of an image in human-readable language, using complete sentences.
5
+
services: cognitive-services
6
+
author: PatrickFarley
7
+
manager: nitinme
8
+
9
+
ms.service: cognitive-services
10
+
ms.subservice: computer-vision
11
+
ms.topic: conceptual
12
+
ms.date: 07/22/2022
13
+
ms.author: pafarley
14
+
---
15
+
16
+
17
+
# Overview: Generate alt text of images with Image Analysis
18
+
19
+
Grow your customer base by making your products and services more accessible. Generate a description of an image in human-readable language, using complete sentences. Computer Vision's algorithms generate various descriptions based on the objects identified in the image. The descriptions are each evaluated, and a confidence score is generated. You then get the list of descriptions, ordered from highest confidence score to lowest.
20
+
21
+
## Image captioning example
22
+
23
+
The following screenshot is an example of automatically generated alt text for an image.
24
+
25
+
:::image type="content" source="media/use-cases/studio-alt-text.png" alt-text="Screenshot of Vision Studio with a caption generated for an image of cows.":::
26
+
27
+
## Adding alt text to images
28
+
29
+
Alternative text (alt text) is descriptive text that conveys the meaning and context of a visual item in a digital setting, such as on an app or web page. When screen readers such as Microsoft Narrator, JAWS, and NVDA reach digital content with alt text, they read the alt text aloud, allowing people to better understand what is on the screen. Well written, descriptive alt text dramatically reduces ambiguity and improves the user experience. Alt text needs to convey the purpose and meaning of an image, which requires understanding and interpretation in addition to object detection.
30
+
31
+
## Key features
32
+
33
+
- Human-readable captions with confidence: Generate a description of an entire image in human-readable language, using complete sentences. Computer Vision's algorithms generate various descriptions based on the objects identified in the image. The descriptions are each evaluated and a confidence score generated. A list is then returned ordered from highest confidence score to lowest.
34
+
- Multiple captions available: You can specify the number of possible captions it will generate and choose the one that works best for your business.
35
+
36
+
## Benefits for your business
37
+
38
+
Accessibility is the design of products, devices, services, vehicles, or environments to be usable by people with disabilities. Accessibility has been an increasingly important topic in product and service development as it makes digital experiences available to more people in the world. Accessibility will boost your business in multiple ways:
39
+
40
+
-**Improve user experience**: by adding alt text, you make the information in images available to users who are blind or have low vision, as well as users who can't load the images due to internet connection.
41
+
-**Make images more discoverable and searchable**: adding captions and tags to images will help search engine crawlers find your images and rank them higher in search results.
42
+
-**Meet legal compliance**: there may potentially be legal repercussions for business that aren't accessibility-compliant. Building products with higher accessibility helps the business avoid potential risks of legal action now and in the future.
43
+
44
+
45
+
46
+
## Next Steps
47
+
The following tutorial provides a complete solution to generate alternative text of images on web applications.
48
+
49
+
> [!div class="nextstepaction"]
50
+
> [Tutorial: Use Computer Vision to generate image metadata](./Tutorials/storage-lab-tutorial.md)
title: "Overview: Monitor dwell time in front of displays"
3
+
titleSuffix: Azure Cognitive Services
4
+
description: Spatial Analysis can provide real-time information about how long customers spend in front of a display in a retail store.
5
+
services: cognitive-services
6
+
author: PatrickFarley
7
+
manager: nitinme
8
+
9
+
ms.service: cognitive-services
10
+
ms.subservice: computer-vision
11
+
ms.topic: conceptual
12
+
ms.date: 07/22/2022
13
+
ms.author: pafarley
14
+
---
15
+
16
+
# Overview: Monitor dwell time in front of displays with Spatial Analysis
17
+
18
+
Spatial Analysis can provide real-time information about how long customers spend in front of a display in a retail store. The service monitors the length of time customers spend in a zone you specif. You can use this information to track customer engagement with promotions/displays within a store or understand customers' preference toward specific products.
19
+
20
+
:::image type="content" source="media/use-cases/dwell-time.jpg" alt-text="Photo of a person in a warehouse with stacks of boxes.":::
21
+
22
+
## Key features
23
+
24
+
You can deploy an end-to-end solution to track customers' time spent in front of displays.
25
+
-**Customize areas to monitor**: with Spatial Analysis, you can customize any space to monitor. It can be spaces in front of promotions or spaces in front of displays.
26
+
-**Count individual entrances**: you can measure how long people spend within these spaces.
27
+
-**Customize models to differentiate between shoppers and employees**: with the customization options offered, you can train and deploy models to differentiate shoppers and employees so that you only generate data about customer behavior.
28
+
29
+
:::image type="content" source="media/use-cases/dwell-time-2.png" alt-text="A camera feed of a store with rectangles drawn around the people.":::
30
+
31
+
## Benefits for your business
32
+
33
+
In-store dwell time is an important metric to help store managers track customers' behaviors and generate insights about customers to make data-driven decisions. Tracking in-store dwell time will help store managers in the following ways:
34
+
***Generate data-driven insights about customers**: by tracking and analyzing the time customers spend in front of specific displays, you can understand how and why customers react to your displays.
35
+
***Collect real-time customer reactions**: instead of sending and receiving feedback from customers, you have access to customer interaction with displays in real time.
36
+
***Measure display/promotion effectiveness**: You can track, analyze and compare footfall and time spent by your customers to measure the effectiveness of promotions or displays.
37
+
***Manage inventory**: You can track and monitor changes in foot traffic and time spent by your customers to modify shift plans and inventory.
38
+
***Make data-driven decisions**: with data about your customers, you're empowered to make decisions that improve customer experience and satisfy customer needs.
39
+
40
+
41
+
## Next steps
42
+
43
+
Get started using Spatial Analysis in a container.
44
+
45
+
> [!div class="nextstepaction"]
46
+
> [Install and run the Spatial Analysis container](./spatial-analysis-container.md)
0 commit comments