What is the Deep Indexer?
Quickly Identify and Organize Unknown Faces and Voices in Your Media
The Deep Indexer is an AI-powered tool that helps you index and identify faces and voices in video, audio, and image files, even if the person or voice is unknown at first. This tool is especially valuable for managing large media archives, where identifying individuals or tracking specific voices across various assets is time-consuming and difficult. It uses clustering techniques to assign unique IDs to each face or voice, making it easier to organize and analyze your media.
Note:
The Deep Indexer is currently focused on file-based analysis, but in the coming months modules will also be available for live resources, which will be integrated into the Deep Live Hub.
The Deep Model Customizer incorporates two modules:
- Face Indexing:
The Deep Indexer can detect and cluster faces across all your media assets, assigning each one a unique ID. This allows you to identify individuals even if they weren’t known when the media was first analyzed. - Voice Indexing:
The Deep Indexer can detect and cluster voices across all your media assets, assigning each one a unique ID. This allows you to identify the voices even in previously analysed assets.
Use Cases:
- Media Companies: Using the Deep Indexer enables you to create a reverse search: Upload an image or voice sample to find all matching instances across your media library.
Manual Relabelling:
After the Deep Indexer has indexed unknown individuals or voices, you can manually rename the unique IDs, ensuring consistent labelling across all future media.