From 2c3730c0c6f1fa411d9caa8398f84cdca41d59e5 Mon Sep 17 00:00:00 2001 From: "Bujewski, Tomasz" Date: Wed, 18 Mar 2026 17:24:57 +0100 Subject: [PATCH 01/12] [DOCS] Update release notes for version 2026.0 with new features, enhancements, and deprecations --- docs/source/release-notes.md | 44 ++++++++++++++++++++++++++---------- 1 file changed, 32 insertions(+), 12 deletions(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index fc3f054a..95bc3bb3 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -57,35 +57,42 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b | YOLOv26 model support| Added converters and post-processing for YOLOv26 models, including oriented bounding box (OBB) support and FP16/FP32 GPU inference. Added YOLOv26 to supported models in samples.| | RT-DETR model support | Added RT-DETR support with dedicated converter implementation for real-time detection transformer models. | | HuggingFace ViT classifier support | Added HuggingFace Vision Transformer (ViT) classifier config parser for inference.| -| gvawatermark enhancements| Major enhancements to the gvawatermark element: display configuration options (thickness, color index), text background support, inclusive/exclusive label filtering, additional font support, average FPS info overlay, and visual documentation.| | Batch timeout for inference elements | Added batch-timeout parameter to inference elements, allowing control over batching wait time.| -| Reference timestamp metadata|Added reference timestamp meta extraction to gvametaconvert element.| -| Custom GstAnalytics data | Enabled custom code to add GstAnalytics data outside of DL Streamer components.| -| Latency tracer multi-source/sink support| Extended latency_tracer to support multiple sources and multiple sinks.| -| Detection anomaly converter | Refactored and enhanced anomaly logic in DetectionAnomalyConverter.| -| FP32 precision in BoxesLabelsConverter | Added FP32 precision support in BoxesLabelsConverter label parsing.| -| Bounding box validation | Added extra validation of bounding boxes to improve robustness.| -| OpenCV tensor compression for all devices | Use OpenCV tensor compression for all inference devices, yielding best performance across CPU/GPU/NPU.| -| Model API refactoring | Moved Model API parser to separate files; added conversion from third-party metadata to Model API.| | VLM Alerts sample (Python) | New Python sample for VLM-based alerts with displaying results on produced video. | | Smart NVR sample (Python) | New Python sample for Smart NVR with prototype elements. | | ONVIF Camera Discovery sample | New sample demonstrating ONVIF camera discovery and DL Streamer pipeline launcher. | | Face detection & age classification sample | New Python sample for face detection and age classification using HuggingFace models. | | Open-vocabulary object detection sample | New Python sample with open-vocabulary prompt for object detection. | +| DL Streamer + DeepStream coexistence sample | New sample demonstrating DL Streamer and DeepStream working in one system. | | RealSense element usage sample | New sample demonstrating gvarealsense element usage. | -| DL Streamer + DeepStream concurrent sample | New sample demonstrating concurrent DL Streamer and DeepStream usage. | + + +**Updated in this Release** + +| Title | High-level description | +|---|---| +| gvawatermark enhancements| Major enhancements to the gvawatermark element: display configuration options (thickness, color index), text background support, inclusive/exclusive label filtering, additional font support, average FPS info overlay, and visual documentation.| +| DLS Optimizer enhancements | Optimizer refactored with multi-stream optimization, cross-stream batching, improved FPS reporting, and device selection improvements. | +| gvametaconvert enhancements| Added reference timestamp meta extraction to gvametaconvert element.| +| Custom GstAnalytics data | Enabled custom code to add GstAnalytics data outside of DL Streamer components.| +| Latency tracer multi-source/sink support| Extended latency_tracer to support multiple sources and multiple sinks.| +| Detection anomaly converter | Refactored and enhanced anomaly logic in DetectionAnomalyConverter.| +| FP32 precision in BoxesLabelsConverter | Added FP32 precision support in BoxesLabelsConverter label parsing.| +| Bounding box validation | Added extra validation of bounding boxes to improve robustness.| +| OpenCV tensor compression for all devices | Use OpenCV tensor compression for all inference devices, yielding best performance across CPU/GPU/NPU.| +| Model API refactoring | Moved Model API parser to separate files; added conversion from third-party metadata to Model API.| | Python samples overview | Added overview section for Python samples; updated READMEs. | +Added functional tests and unit tests. | | Windows: GPU inference with D3D11 | Added support for GPU inference on Windows using D3D11. | | Windows: gvapython support | Added Windows support for gvapython element and gstgva Python bindings. | | Windows: enhanced build & setup | Enhanced Windows build/setup scripts, added remove script, Visual C++ runtime handling, and JSON output for Windows samples. | | Windows: CI integration | Enabled Windows tests in GitHub Actions workflow, model downloads on Windows. | -| DL Optimizer enhancements | Optimizer refactored with multi-stream optimization, cross-stream batching, improved FPS reporting, and device selection improvements. Added functional tests and unit tests. | -| CI: Zizmor security scanning | Added Zizmor GitHub Actions security scanner. | | Library consolidation | Merged gvawatermark3d, gvadeskew, gvamotiondetect, and gvagenai into the gstvideoanalytics library. | | OpenVINO update | Update to OpenVINO 2026.0.0. | | NPU driver update | Update to NPU driver version 1.30. | | RealSense update | Update to Intel RealSense SDK 2.57.5. | | Model download script improvements | Simplified YOLO model download script, enhanced INT8 quantization, refactored YOLOv8+ export/quantize, added model validation. | +| CI: Zizmor security scanning | Added Zizmor GitHub Actions security scanner. | **Fixed** @@ -120,6 +127,19 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b |---|---| | Preview Architecture 2.0 Samples | Preview Arch 2.0 samples have known issues with inference results. | + +## Deprecation And Support +List of the features and components to be deprecated in the future. + +| Feature | End of Support Date | Replacement strategy | +|---|---|---| +| OMZ models | end of Q3'2026 | Use HuggingFace, Ultralytics, TIMM | +| Model-proc-file| end of Q3'2026 | ModelAPI (common with Geti) | +| GstVideoRegionOfInterest meta + Intel extensions | end of Q4'2026 | GstAnalyticsMtd (aka new Collabora meta) | +| FFMpeg integration samples | end of Q2'2026 | no replacement | +| Architecture 2.0 elements | end of Q4'2026 (?) need to take valuable elements/ideas and deprecate others | + + ## Legal Information * GStreamer is an open source framework licensed under LGPL. From aee21ab12d53b65533d3f6dad0c4888aee3312c7 Mon Sep 17 00:00:00 2001 From: "Bujewski, Tomasz" Date: Wed, 18 Mar 2026 17:25:03 +0100 Subject: [PATCH 02/12] [DOCS] Remove deprecated architecture 2.0 elements from release notes --- docs/source/release-notes.md | 1 - 1 file changed, 1 deletion(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index 95bc3bb3..9904b624 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -137,7 +137,6 @@ List of the features and components to be deprecated in the future. | Model-proc-file| end of Q3'2026 | ModelAPI (common with Geti) | | GstVideoRegionOfInterest meta + Intel extensions | end of Q4'2026 | GstAnalyticsMtd (aka new Collabora meta) | | FFMpeg integration samples | end of Q2'2026 | no replacement | -| Architecture 2.0 elements | end of Q4'2026 (?) need to take valuable elements/ideas and deprecate others | ## Legal Information From f6b46a436c7d86996e1b4c9d467258be84276ede Mon Sep 17 00:00:00 2001 From: "Bujewski, Tomasz" Date: Wed, 18 Mar 2026 17:29:27 +0100 Subject: [PATCH 03/12] [DOCS] Update deprecation timeline for Architecture 2.0 elements in release notes --- docs/source/release-notes.md | 1 + 1 file changed, 1 insertion(+) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index 9904b624..f24de33b 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -136,6 +136,7 @@ List of the features and components to be deprecated in the future. | OMZ models | end of Q3'2026 | Use HuggingFace, Ultralytics, TIMM | | Model-proc-file| end of Q3'2026 | ModelAPI (common with Geti) | | GstVideoRegionOfInterest meta + Intel extensions | end of Q4'2026 | GstAnalyticsMtd (aka new Collabora meta) | +| Architecture 2.0 elements | end of Q4'2026 | need to take valuable elements/ideas and deprecate others | | FFMpeg integration samples | end of Q2'2026 | no replacement | From 4a1ce01d41e8bf21e8cb7149832a9760299fb774 Mon Sep 17 00:00:00 2001 From: Tomasz Bujewski Date: Thu, 19 Mar 2026 09:39:26 +0100 Subject: [PATCH 04/12] fix per review --- docs/source/release-notes.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index f24de33b..31fd8658 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -54,7 +54,7 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b |---|---| | 3D elements (g3dradarprocess, g3dlidarparse) | New 3D plugin support with g3dradarprocess element for radar data processing and g3dlidarparse element for lidar data parsing, enabling 3D analytics pipelines.| | FPS throttle element (gvafpsthrottle) | New element to throttle the frame rate of a pipeline to a specified FPS value. | -| YOLOv26 model support| Added converters and post-processing for YOLOv26 models, including oriented bounding box (OBB) support and FP16/FP32 GPU inference. Added YOLOv26 to supported models in samples.| +| YOLOv26 model support| Added converters and post-processing for YOLOv26 models, including oriented bounding box (OBB) support and INT8 GPU inference. Added YOLOv26 to supported models in samples.| | RT-DETR model support | Added RT-DETR support with dedicated converter implementation for real-time detection transformer models. | | HuggingFace ViT classifier support | Added HuggingFace Vision Transformer (ViT) classifier config parser for inference.| | Batch timeout for inference elements | Added batch-timeout parameter to inference elements, allowing control over batching wait time.| From 24a8ad6a953707c5747252d36f5f7b08d6a784c6 Mon Sep 17 00:00:00 2001 From: Tomasz Bujewski Date: Thu, 19 Mar 2026 12:14:19 +0100 Subject: [PATCH 05/12] Apply suggestion from @oonyshch Co-authored-by: Olena Onyshchenko --- docs/source/release-notes.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index 31fd8658..d810f484 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -82,7 +82,7 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b | OpenCV tensor compression for all devices | Use OpenCV tensor compression for all inference devices, yielding best performance across CPU/GPU/NPU.| | Model API refactoring | Moved Model API parser to separate files; added conversion from third-party metadata to Model API.| | Python samples overview | Added overview section for Python samples; updated READMEs. | -Added functional tests and unit tests. | +| Tests | Expanded coverage of functional and unit tests. | | Windows: GPU inference with D3D11 | Added support for GPU inference on Windows using D3D11. | | Windows: gvapython support | Added Windows support for gvapython element and gstgva Python bindings. | | Windows: enhanced build & setup | Enhanced Windows build/setup scripts, added remove script, Visual C++ runtime handling, and JSON output for Windows samples. | From 8f453e1d0423b730c8b17ce9b33fe915e9dc6c4e Mon Sep 17 00:00:00 2001 From: Tomasz Bujewski Date: Thu, 19 Mar 2026 12:31:59 +0100 Subject: [PATCH 06/12] review fixes #2 --- docs/source/release-notes.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index d810f484..8e3bc6fe 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -55,11 +55,11 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b | 3D elements (g3dradarprocess, g3dlidarparse) | New 3D plugin support with g3dradarprocess element for radar data processing and g3dlidarparse element for lidar data parsing, enabling 3D analytics pipelines.| | FPS throttle element (gvafpsthrottle) | New element to throttle the frame rate of a pipeline to a specified FPS value. | | YOLOv26 model support| Added converters and post-processing for YOLOv26 models, including oriented bounding box (OBB) support and INT8 GPU inference. Added YOLOv26 to supported models in samples.| -| RT-DETR model support | Added RT-DETR support with dedicated converter implementation for real-time detection transformer models. | +| RT-DETR model support | Added RT-DETR support implementation for real-time detection transformer models. | | HuggingFace ViT classifier support | Added HuggingFace Vision Transformer (ViT) classifier config parser for inference.| | Batch timeout for inference elements | Added batch-timeout parameter to inference elements, allowing control over batching wait time.| | VLM Alerts sample (Python) | New Python sample for VLM-based alerts with displaying results on produced video. | -| Smart NVR sample (Python) | New Python sample for Smart NVR with prototype elements. | +| Smart NVR sample (Python) | New Python sample for Smart NVR with added custom analytics logic (gvaAnalytics) and custom storage (gvaRecorder) elements. | | ONVIF Camera Discovery sample | New sample demonstrating ONVIF camera discovery and DL Streamer pipeline launcher. | | Face detection & age classification sample | New Python sample for face detection and age classification using HuggingFace models. | | Open-vocabulary object detection sample | New Python sample with open-vocabulary prompt for object detection. | @@ -80,7 +80,7 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b | FP32 precision in BoxesLabelsConverter | Added FP32 precision support in BoxesLabelsConverter label parsing.| | Bounding box validation | Added extra validation of bounding boxes to improve robustness.| | OpenCV tensor compression for all devices | Use OpenCV tensor compression for all inference devices, yielding best performance across CPU/GPU/NPU.| -| Model API refactoring | Moved Model API parser to separate files; added conversion from third-party metadata to Model API.| +| Model API refactoring | Moved Model API parser to separate files; added conversion from Ultralytics and HuggingFace metadata to Model API.| | Python samples overview | Added overview section for Python samples; updated READMEs. | | Tests | Expanded coverage of functional and unit tests. | | Windows: GPU inference with D3D11 | Added support for GPU inference on Windows using D3D11. | @@ -128,15 +128,15 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b | Preview Architecture 2.0 Samples | Preview Arch 2.0 samples have known issues with inference results. | -## Deprecation And Support +## Legacy Features and Deprecation Timeline List of the features and components to be deprecated in the future. | Feature | End of Support Date | Replacement strategy | |---|---|---| | OMZ models | end of Q3'2026 | Use HuggingFace, Ultralytics, TIMM | | Model-proc-file| end of Q3'2026 | ModelAPI (common with Geti) | -| GstVideoRegionOfInterest meta + Intel extensions | end of Q4'2026 | GstAnalyticsMtd (aka new Collabora meta) | -| Architecture 2.0 elements | end of Q4'2026 | need to take valuable elements/ideas and deprecate others | +| GstVideoRegionOfInterest meta + Intel extensions | end of Q4'2026 | usage of GstAnalyticsMtd | +| Architecture 2.0 elements | end of Q4'2026 | Preserve essential components and ideas | | FFMpeg integration samples | end of Q2'2026 | no replacement | From dd8e9ceacbcbf6030481b72411329ca684a8256e Mon Sep 17 00:00:00 2001 From: Tomasz Bujewski Date: Thu, 19 Mar 2026 12:55:31 +0100 Subject: [PATCH 07/12] review fixes #3 --- docs/source/release-notes.md | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index 8e3bc6fe..a92307f4 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -4,7 +4,8 @@ ## Key highlights: * New elements: gvafpsthrottle, g3dradarprocess, g3dlidarparse -* New model support: YOLOv26 (including OBB), RT-DETR, HuggingFace ViT +* New model support: YOLOv26, YOLO-E, RT-DETR, HuggingFace ViT +* Streamlined integration with Ultralytics and HuggingFace model hubs * gvawatermark overhaul: object bluring, text backgrounds, label filtering, extra fonts, thickness/color options, FPS overlay * Inference enhancements: batch timeout, OpenCV tensor compression for all devices, FP32 precision, custom GstAnalytics data API * Windows platform: GPU inference via D3D11, gvapython support, CI integration, build/setup improvements @@ -64,6 +65,7 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b | Face detection & age classification sample | New Python sample for face detection and age classification using HuggingFace models. | | Open-vocabulary object detection sample | New Python sample with open-vocabulary prompt for object detection. | | DL Streamer + DeepStream coexistence sample | New sample demonstrating DL Streamer and DeepStream working in one system. | +| Motion detect sample (Windows) | New sample demonstrating DL Streamer gvamotiondetect functionality | | RealSense element usage sample | New sample demonstrating gvarealsense element usage. | @@ -74,7 +76,7 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b | gvawatermark enhancements| Major enhancements to the gvawatermark element: display configuration options (thickness, color index), text background support, inclusive/exclusive label filtering, additional font support, average FPS info overlay, and visual documentation.| | DLS Optimizer enhancements | Optimizer refactored with multi-stream optimization, cross-stream batching, improved FPS reporting, and device selection improvements. | | gvametaconvert enhancements| Added reference timestamp meta extraction to gvametaconvert element.| -| Custom GstAnalytics data | Enabled custom code to add GstAnalytics data outside of DL Streamer components.| +| ROI object construction enhacement| For existing GstAnalyticsODMtd only, creates GstVideoRegionOfInterestMeta until full GstAnalytics migration| | Latency tracer multi-source/sink support| Extended latency_tracer to support multiple sources and multiple sinks.| | Detection anomaly converter | Refactored and enhanced anomaly logic in DetectionAnomalyConverter.| | FP32 precision in BoxesLabelsConverter | Added FP32 precision support in BoxesLabelsConverter label parsing.| From 21227d8dd39f1cc820020c99b942173fff9f29b9 Mon Sep 17 00:00:00 2001 From: Tomasz Bujewski Date: Thu, 19 Mar 2026 12:57:45 +0100 Subject: [PATCH 08/12] Apply suggestion from @oonyshch Co-authored-by: Olena Onyshchenko --- docs/source/release-notes.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index a92307f4..399a8339 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -59,7 +59,7 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b | RT-DETR model support | Added RT-DETR support implementation for real-time detection transformer models. | | HuggingFace ViT classifier support | Added HuggingFace Vision Transformer (ViT) classifier config parser for inference.| | Batch timeout for inference elements | Added batch-timeout parameter to inference elements, allowing control over batching wait time.| -| VLM Alerts sample (Python) | New Python sample for VLM-based alerts with displaying results on produced video. | +| VLM Alerts sample | New Python sample for VLM-based alerts with displaying results on produced video. | | Smart NVR sample (Python) | New Python sample for Smart NVR with added custom analytics logic (gvaAnalytics) and custom storage (gvaRecorder) elements. | | ONVIF Camera Discovery sample | New sample demonstrating ONVIF camera discovery and DL Streamer pipeline launcher. | | Face detection & age classification sample | New Python sample for face detection and age classification using HuggingFace models. | From 1d3e7f1f8faf07adb07dff7292cef5a0e364d253 Mon Sep 17 00:00:00 2001 From: Tomasz Bujewski Date: Thu, 19 Mar 2026 12:57:53 +0100 Subject: [PATCH 09/12] Apply suggestion from @oonyshch Co-authored-by: Olena Onyshchenko --- docs/source/release-notes.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index 399a8339..b1fe8c55 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -61,7 +61,7 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b | Batch timeout for inference elements | Added batch-timeout parameter to inference elements, allowing control over batching wait time.| | VLM Alerts sample | New Python sample for VLM-based alerts with displaying results on produced video. | | Smart NVR sample (Python) | New Python sample for Smart NVR with added custom analytics logic (gvaAnalytics) and custom storage (gvaRecorder) elements. | -| ONVIF Camera Discovery sample | New sample demonstrating ONVIF camera discovery and DL Streamer pipeline launcher. | +| ONVIF Camera Discovery sample | New Python sample demonstrating ONVIF camera discovery and DL Streamer pipeline launcher. | | Face detection & age classification sample | New Python sample for face detection and age classification using HuggingFace models. | | Open-vocabulary object detection sample | New Python sample with open-vocabulary prompt for object detection. | | DL Streamer + DeepStream coexistence sample | New sample demonstrating DL Streamer and DeepStream working in one system. | From 87994480f9f4ff5e92f0e501b006804332d40fec Mon Sep 17 00:00:00 2001 From: Tomasz Bujewski Date: Thu, 19 Mar 2026 12:58:01 +0100 Subject: [PATCH 10/12] Apply suggestion from @oonyshch Co-authored-by: Olena Onyshchenko --- docs/source/release-notes.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index b1fe8c55..64114659 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -60,7 +60,7 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b | HuggingFace ViT classifier support | Added HuggingFace Vision Transformer (ViT) classifier config parser for inference.| | Batch timeout for inference elements | Added batch-timeout parameter to inference elements, allowing control over batching wait time.| | VLM Alerts sample | New Python sample for VLM-based alerts with displaying results on produced video. | -| Smart NVR sample (Python) | New Python sample for Smart NVR with added custom analytics logic (gvaAnalytics) and custom storage (gvaRecorder) elements. | +| Smart NVR sample | New Python sample for Smart NVR with added custom analytics logic (gvaAnalytics) and custom storage (gvaRecorder) elements. | | ONVIF Camera Discovery sample | New Python sample demonstrating ONVIF camera discovery and DL Streamer pipeline launcher. | | Face detection & age classification sample | New Python sample for face detection and age classification using HuggingFace models. | | Open-vocabulary object detection sample | New Python sample with open-vocabulary prompt for object detection. | From 433e38654188bd1b8adda3b8973853443de46dff Mon Sep 17 00:00:00 2001 From: Tomasz Bujewski Date: Thu, 19 Mar 2026 13:08:38 +0100 Subject: [PATCH 11/12] review fixes #4 --- docs/source/release-notes.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index 64114659..0434453a 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -6,8 +6,9 @@ * New elements: gvafpsthrottle, g3dradarprocess, g3dlidarparse * New model support: YOLOv26, YOLO-E, RT-DETR, HuggingFace ViT * Streamlined integration with Ultralytics and HuggingFace model hubs +* GstAnalytics metadata support: DLStreamer supports GstAnalytics metadata for object detection, classification, tracking and adds custom GstAnalytics extension for keypoints * gvawatermark overhaul: object bluring, text backgrounds, label filtering, extra fonts, thickness/color options, FPS overlay -* Inference enhancements: batch timeout, OpenCV tensor compression for all devices, FP32 precision, custom GstAnalytics data API +* Inference enhancements: batch timeout, OpenCV tensor compression for all devices * Windows platform: GPU inference via D3D11, gvapython support, CI integration, build/setup improvements * New Python samples: VLM Alerts, Smart NVR, ONVIF Discovery, face detection/age classification, open-vocabulary detection, RealSense, DL Streamer + DeepStream * Optimizer: multi-stream optimization, cross-stream batching, device selection, refactored with tests From 38436ba594c630cc3db5d623df4fd7a4d1aaf747 Mon Sep 17 00:00:00 2001 From: Tomasz Bujewski Date: Fri, 20 Mar 2026 09:58:10 +0100 Subject: [PATCH 12/12] gvametaconvert change description update --- docs/source/release-notes.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md index 0434453a..a90b6a89 100644 --- a/docs/source/release-notes.md +++ b/docs/source/release-notes.md @@ -76,7 +76,7 @@ For installing Pipeline Framework with the prebuilt binaries or Docker\* or to b |---|---| | gvawatermark enhancements| Major enhancements to the gvawatermark element: display configuration options (thickness, color index), text background support, inclusive/exclusive label filtering, additional font support, average FPS info overlay, and visual documentation.| | DLS Optimizer enhancements | Optimizer refactored with multi-stream optimization, cross-stream batching, improved FPS reporting, and device selection improvements. | -| gvametaconvert enhancements| Added reference timestamp meta extraction to gvametaconvert element.| +| gvametaconvert enhancements| Added reference NTP timestamp from RTCP sender meta extraction to gvametaconvert element| | ROI object construction enhacement| For existing GstAnalyticsODMtd only, creates GstVideoRegionOfInterestMeta until full GstAnalytics migration| | Latency tracer multi-source/sink support| Extended latency_tracer to support multiple sources and multiple sinks.| | Detection anomaly converter | Refactored and enhanced anomaly logic in DetectionAnomalyConverter.|