Release Notes: Loitering Detection#

Version 2.0.0 (Yet to be released)#

New:

  • Zone presence and dwell-time are now computed natively in the pipeline via the gvaanalytics element, with the loitering_watermark element rendering the dwell-time/status overlay — no external low-code flow engine required.

  • Added a lightweight mqtt-table-flattener sidecar that reshapes per-frame MQTT metadata into one-row-per-object summaries so the Grafana status table renders correctly.

  • Zones are now defined per stream as JSON polygon/circle configs, so adding or editing zones requires no pipeline or dashboard changes.

Removed:

  • Removed the Node-RED service and flow; its zone/dwell-time logic and Grafana data reshaping are now handled natively as described above.

Improved:

  • Consolidated the Grafana dashboard’s per-stream tables into a single status table.

Version 1.6.0#

Release Date: September 10, 2026

New:

  • Fixed coturn server configuration for WebRTC relay.

  • Fixed Grafana MQTT datasource version to avoid errors with the latest version.

Improved:

  • Improved the security context of Grafana and Node-RED containers.

  • Consumed the latest DL Streamer version 2026.2.0.

Version 1.5.0#

New:

  • Qualified on the Intel® Core™ Series 3 processor “Wildcat Lake” (WCL).

  • Deprecated EMF deployment packages.

Improved:

  • Optimized the latency for GPU and NPU workloads.

Version 1.4.0#

New:

  • Qualified on the Intel® Core™ Ultra Series 3 processor.

Improved:

  • Consumed the latest DL Streamer Pipeline Server 2026.0.0 image. Ubuntu24 variant of the image is the default now.

  • Optimized the latency for GPU and NPU workloads.

Version 1.3.0#

New:

  • Consumed latest DL Streamer Pipeline Server version 2025.2

  • Introduced nginx server as reverse proxy and TLS

  • Optimized pipelines and quantized model from FP32 to FP16.

For more details on the new features and improvements, refer to the overview.

Version 1.2#

Release Date: March 28, 2025

Features:

  • Enhanced Object Detection Models: Improved accuracy and performance of pre-trained deep learning models for object detection and classification.

  • Advanced Analytics Dashboard: Grafana dashboards with additional metrics and visualizations for better monitoring and analysis.

  • Customizable Alerting System: Integration with Node-RED for setting up custom alerts and notifications based on specific events and thresholds.

  • Expanded Device Support: Added compatibility with a wider range of cameras and video input devices.

  • Optimized Edge Processing: Performance enhancements for real-time video processing on edge devices, reducing latency and resource usage.