Photo Text / Optical Character Recognition (text inside the image) allows you to search for keywords inside the images. This can be very useful to detect the presence of your brand mentions where your brand's name is not mentioned in the message but an image is shared with your brand’s name in it. Also, the photo text does not have to be your brand’s name, it can be any combination of letters and we will search for it.

Photo Text is offered as part of the Visual Insights package; it uses AI to detect letters written as part of the query (to create Photo Text queries refer to OCR Topic Setup ) in images.

What is an OCR Mention?

OCR stands for Optical Character Recognition. It is a technology that recognizes text within a digital image.

There are many messages which do not have your brand logo but have names directly written in text format like memes, screenshots, and edited pictures. The name of your brand is indirectly linked with these mentions and affects the brand's reputation.

Key use cases

• Image and video analysis: OCR can be used to extract text from images (memes, shared posts) and videos (thumbnails) to help in tasks such as search Brand or Product relevant terms inside images, content moderation, and automated tagging.

• As a Content moderator, I am responsible for reviewing and flagging inappropriate or offensive content from social media platforms, websites, and other online platforms related to my brand / services or products.

• As an Analyst I want to understand what is the sentiment around my brand which is shared across social media in terms of memes/ photo updates or videos at various places and how are people responding to it.

How will it look?

Example If you were to create a Photo Text query with the word "sprinklr" these mentions will get fetched as they have the word "sprinklr" mentioned in them.

All the mentions brought in through the Photo Text will be automatically enriched with all Visual Dimensions.

Limitations

• Photo Text is case sensitive.