We need the ability for Bynder to process video assets and perform the following:
- people recognition (face detection), integrated with people already detected in image assets
- object detection - eg there is a dog or frog in this video
- text detection - eg there is a sign on a building, the AI should OCR that sign’s text
All of these are current features in your competitor’s products.
I should be able to search for “cat” and Bynder should recognize that object in image and video assets.
Ideally the system would allow me to perform a natural language search like “an arial shot of a city with skyscrapers, blue sky and mountains”. Or “an image with a dog and a cat”.
These searches must be cross domain - so finds in image assets, video assets, and other document (eg PDF).