How to use Camera Trainer (Bosch IVA- Intelligent Video Analytics)?

Based on examples of target objects and non-target objects, the Camera Trainer uses machine learning to allow the user to define objects of interest and generate detectors for them.

In contrast to the moving objects detected in general by IVA 7.10, Camera Trainer allows detection of both moving and non-moving objects, separates, and immediately classifies them.

Training can be done both on live video as well as on recordings available through the respective camera. The resulting detectors can be down- and uploaded for distribution to other cameras.

The attached White paper describes in-depth what Camera Trainer is, how it works, what to use it for and what not, and how to configure it. The configuration is also shown in the following how-to videos:

Whitepaper_IVA_CameraTrainer_Technical_Note.pdf

Step-by-step guide

Setup Camera Trainer

Camera learning phase with recorded video

Configure Camera Trainer Task

Backup Camera Trainer Configuration - Method 1

Backup Camera Trainer Configuration - Method 2

Setup VCA Masking