Appearance
What's New in 2.7
Welcome to FORXAI Video Vision 2.7!
Summary
This minor release bring improvements to machine learning capabilities and extends business logic available to configure Flows.
Review this section for more information about ability to adjust polygon dimensions based on two different coordinate systems and ability to store detection data between multiple flow iterations.
Support for new neural network - YOLO object detection was added, enhancing capabilities of platform to run new models.
See information about hotfix updates:
Major Features
Major features introduced in this release are described below. This release includes other changes related to fine-tuning FVV capabilities, bugfixes, UX and other improvements which are omitted in this article.
New nodes
Adjust Polygons to Original Image
Get shared memory and Set shared memory
The 'Get Shared Memory' and 'Set Shared Memory' nodes work together to temporarily store and retrieve data.
The 'Set Shared Memory' node saves input data using a defined 'key,' allowing the 'Get Shared Memory' node to later access this value by referencing the same key. This connection enables data continuity and data transfer between nodes and flows.
YOLO object detection
Added support for object detection using the YOLO neural network.
Fixed Issues
Fixed an issue where image snapshots would disappear from the customer dashboard upon receiving a new image.
Fixed an issue where some Prometheus metrics were not working correctly or were unavailable.
Compatibility
The Video Vision platform was tested on Ubuntu Server 22.04.5 LTS. No breaking changes were identified.
This version powers minor release of Forxai Mirror v2.3.0.
v2.7.1
Fix troubleshooting (Prometheus) metrics not working properly.
This version powers hotfix release of Forxai Mirror v2.3.1.