You have a bot that identifies the brand names of products in images of supermarket shelves.
Which service does the bot use?
You have a bot that identifies the brand names of products in images of supermarket shelves.
Which service does the bot use?
Computer Vision Image Analysis capabilities are used to analyze images to detect and identify objects and text within them, making it suitable for recognizing brand names in images of supermarket shelves. This service is specifically designed for tasks like identifying commercial brands in images or video by leveraging a database of global logos.
Can't delete my previous comment. I guess I was wrong. The correct answer should be B. https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/concept-brand-detection
B is correct.
For a bot that identifies brand names of products in images of supermarket shelves, the service used is: B. Computer Vision Image Analysis capabilities Computer Vision can analyze images to detect and identify objects and text, making it ideal for recognizing brand names in images. Incorrect Options: A. AI enrichment for Azure Search capabilities: This is more for enhancing search capabilities with AI, not specifically for image analysis. C. Custom Vision Image Classification capabilities: While it can classify images, Custom Vision requires specific training for each brand, which might not be practical for a wide range of supermarket products. D. Language Understanding capabilities: This service is used for understanding spoken or written language, not for image analysis.
Yes https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/concept-brand-detection
B is correct
C is the Correct one
It feels like it should be OCR but the closest match appears to be Image analysis it seems
Actually, the real service to be used in this context is OCR, if we assume that the brand name (and not just the logos) are visible on the packages. Training a custom Computer Vision model for this task is a bit of an overkill.
ChatGPT
Answer B. Computer Vision Image Analysis capabilities. Computer Vision is the domain of AI and machine learning that focuses on understanding and analyzing images and video content, making it well-suited for tasks like identifying brand names within images.
Custom Vision is a part of Azure’s Cognitive Services that allows you to build and refine your own custom image classification models. It can be trained to recognize various visual elements, including brand logos or specific products on supermarket shelves. While Computer Vision Image Analysis capabilities can analyze and describe an image with tags, it’s not specifically designed for custom tasks like identifying specific brand names in images.
Supermarkets have tens of thousands, if not hundred, of products. I believe Microsoft can identify big brands, like Tesla, Coca Cola, etc. They might have imported thousands of logos, but I don't think they can cover everything in a supermarket.
The answer is B, Computer Vision. Brand detection is a specialized mode of object detection that uses a database of thousands of global logos to identify commercial brands in images or video. You can use this feature, for example, to discover which brands are most popular on social media or most prevalent in media product placement. https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/concept-brand-detection
https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/concept-brand-detection
Computer Vision does not have the capabilities to recognize custom entities. C is the correct answer
Question is about brands, not custom entities. Correct answer is B.
Custom Vision allows you to train a model to recognize specific brands and objects within images, which is ideal for identifying and classifying various product brands on supermarket shelves.
It should be Custom Vision. Computer Vision can't be trained for all the brands in the world (in multiple countries and regions).
But in combination with the classification it does not make sense. So it must be B or the question is simply wrong...
In fact Computer Vision is trained to recognise thousands of brands present across the globe.
This is Image Analysis and not Image Classification.