PLATFORM OVERVIEW / AI / AI VISION

Cloudinary
AI Vision (Beta)

A specialized AI feature that automates media management, enabling precise, scalable, brand-specific content workflows.

Let’s Chat

Media Intelligence Powered By GenAI

We’ve added the power of GenAI to the Modern DAM. Use simple, image-related queries to find, classify, and moderate images, no matter how vast your asset library may be. Media management and moderation is now on your terms. Literally. 

Cloudinary AI

What is AI Vision?

AI Vision enhances media management by leveraging Gen AI. It utilizes a generative multimodal LLM to interpret and respond to visual content queries and prompts, driving automation of key processes, including content moderation, image classification, and custom tagging. AI Vision helps businesses streamline moderation operations, and improve classification capabilities at scale.

By combining a generative multimodal LLM and our own expertise in image AI, AI Vision interprets and responds to visual content queries and prompts, to automate content moderation, image classification, and custom tagging. AI Vision does what Standard LLM’s  can’t.

AI Vision Use Cases

Cloudinary AI

General Question and Answering

Receive detailed, context-aware answers to questions about your images. AI Vision utilizes generative LLMs to identify objects, scenes, and interpret in-image text  so your media assets are more searchable and better organized. Advanced usage is possible through custom workflows with nuanced prompting, such as identifying and scoring the most relevant images for a product page based on content or scenery.

Sample prompts

Can you describe the setting of this image?

  • AI RESPONSE: The image appears to depict a fitness or gym setting. The background shows various exercise equipment and punching bags, suggesting this is a workout or training facility. The individual in the image is positioned in a crouched stance, likely engaged in some form of exercise or training routine within this gym environment.

Sample prompts

How many people are in this image?

  • AI RESPONSE: The image shows one person, a male individual, in the fitness or gym setting.
Cloudinary AI

Content Moderation and Compliance

AI Vision provides straightforward Yes/No/Unknown responses for quick, accurate brand compliance checks and flag potentially sensitive content. An automated workflow moderates content on the fly while maintaining standards across all platforms without manual effort.

Sample prompts

Does the image feature a celebrity or a public figure?

  • AI RESPONSE: No

Sample prompts

Is there anything in the image that could be considered violent or disturbing?

  • AI RESPONSE: No

Sample prompts

Is the image cropped in a way that facial features such as eyes, nose, or mouth are not visible?

  • AI RESPONSE: No
Cloudinary AI

Custom Taxonomy and Image Classification

Classify images based on their unique taxonomy without needing to train or fine-tune tagging models. By providing a set of tags with specific descriptions, businesses can categorize images according to their branding and organizational needs. Quickly and accurately tag images based on detailed criteria like background color and subject orientation. Demographics can be built into an automated workflow that can analyze images at scale.

Ready to start using Cloudinary?

Sign up for our free plan and start creating stunning visual experiences in minutes.
Sign Up for Free