Use Case - DEEP MODEL CUSTOMIZER

Individual person identification  

How can I recognize less-known or regional people in my footage?

Anyone who has image or video material in their inventory and would like to train a face recognition dataset from it, including broadcasters, publishers, or media archives, among others.

The Problem

In many cases, image and video archives do not only contain known, public persons. This results in the challenge that only certain persons, i.e. persons for whom the AI has already been trained, are recognized. However, it is precisely persons who repeatedly appear in the image and video material that cannot be identified in this way. How can DeepVA help me to recognize people who are important to me individually in my video and image material?

The Solution

With the DeepVA Deep Model Customizer you can easily manage the recognition of individual people in your media material. With the so call datasets, you can not only create classes but also enable, disable, merge or delete personified information. It can be easily determined within each class what should or should not be included in the training. Data sets, as well as image data, can be evaluated, i.e. it can be determined which image data is suitable for successful training and which is not. Faulty or poorly recognizable image data is left out of the equation.

0 %

faster data acquisition

0 %

COST REDUCTION

0 %

faster labelling

What results can be obtained?

An easy-to-manage database in which training data can be structured.

The individual classes (objects, people, buildings) receive a quality control with feedback to the user

You have the option of leaving individual classes (persons) only temporarily for a certain period of time in the system and the system automatically forgets them

Individual person identification  

Function overview

Face Training

In just a few seconds, new faces are trained using our Face Training and can be used in the Recognition Services.

Requirements

Contact us

Do you have any questions?

Enter your email below and we will get back to you as soon as possible.

Related use cases

Take a look at our other use cases

Identification of speakers

How can I recognize new speakers in my media assets in the future?

Read More