Merge branch 'release-2025.2' into mapping_service_helm_chart_deployment

Author: Irakus

autocalibration/src/auto_camera_calibration_controller.py

@@ -72,8 +72,20 @@ def isMapProcessed(self, sceneobj):
 
     @return  True/False
     """
-    return (sceneobj.map_processed < datetime.fromtimestamp(os.path.getmtime(sceneobj.map),
-                                                        tz=timezone(TIMEZONE)))
+    if not sceneobj.map or not getattr(sceneobj.map, 'path', None):
+      return False
+
+    try:
+      map_path = sceneobj.map.path
+      map_mtime = datetime.fromtimestamp(os.path.getmtime(map_path), tz=timezone(TIMEZONE))
+    except (FileNotFoundError, ValueError, TypeError) as e:
+      log.warning(f"Map file missing or invalid for scene {sceneobj.id}: {e}")
+      return False
+
+    if sceneobj.map_processed is None:
+      return False
+
+    return sceneobj.map_processed < map_mtime
 
   def saveToDatabase(self, scene):
     """! Function stores baseapriltag data into db.

autocalibration/src/auto_camera_calibration_model.py

@@ -6,8 +6,8 @@
 from scene_common.timestamp import get_datetime_from_string
 from scene_common.transform import CameraIntrinsics
 
-
 class CalibrationScene:
+  _MEDIA_PATH = "/home/scenescape/SceneScape/media/"
   #FIXME: should be defined in common location accessible to both models and camcalibration
   DEFAULTS = {
     'rotation_x': 0.0,
@@ -48,15 +48,15 @@ def __init__(self, uid, name):
   @classmethod
   def deserialize(cls, data):
     scene = cls(data['uid'], data['name'])
-    scene.map = "/workspace/media/" + data['map'].split('/')[-1] if 'map' in data else scene.DEFAULTS['map']
+    scene.map = cls._MEDIA_PATH + data['map'].split('/')[-1] if 'map' in data else scene.DEFAULTS['map']
     scene.scale = data.get('scale', scene.DEFAULTS['scale'])
     scene.mesh_rotation = data.get('mesh_rotation', scene.mesh_rotation)
     scene.mesh_translation = data.get('mesh_translation', scene.mesh_translation)
     scene.mesh_scale = data.get('mesh_scale', scene.mesh_scale)
     scene.apriltag_size = data.get('apriltag_size', scene.DEFAULTS['apriltag_size'])
     scene.map_processed = get_datetime_from_string(data['map_processed']) if 'map_processed' in data else scene.DEFAULTS['map_processed']
     scene.camera_calibration = data.get('camera_calibration', scene.DEFAULTS['camera_calibration'])
-    scene.polycam_data = "/workspace/media/" + data['polycam_data'].split('/')[-1] if 'polycam_data' in data else scene.DEFAULTS['polycam_data']
+    scene.polycam_data = cls._MEDIA_PATH + data['polycam_data'].split('/')[-1] if 'polycam_data' in data else scene.DEFAULTS['polycam_data']
     scene.number_of_localizations = data.get('number_of_localizations', scene.DEFAULTS['number_of_localizations'])
     scene.global_feature = data.get('global_feature', scene.DEFAULTS['global_feature'])
     scene.local_feature = data.get('local_feature', scene.DEFAULTS['local_feature'])

controller/src/schema/metadata.schema.json

@@ -207,9 +207,9 @@
         },
         "age": {
           "title": "Age",
-          "type": "integer",
+          "type": "string",
           "description": "The age in years of a detection, such as the result of a DL-based age detector.",
-          "examples": [21, 33, 65]
+          "examples": ["21", "33", "65"]
         },
         "hat": {
           "title": "Hat",

dlstreamer-pipeline-server/model_config.json

@@ -5,11 +5,17 @@
       "model": "intel/person-detection-retail-0013/FP32/person-detection-retail-0013.xml",
       "model_proc": "object_detection/person/person-detection-retail-0013.json"
     },
-    "input-format": {
-      "color-space": "BGR"
-    },
     "adapter-params": {
       "metadatagenpolicy": "detectionPolicy"
     }
+  },
+  "reid": {
+    "type": "inference",
+    "params": {
+      "model": "intel/person-reidentification-retail-0277/FP32/person-reidentification-retail-0277.xml"
+    },
+    "adapter-params": {
+      "metadatagenpolicy": "reidPolicy"
+    }
   }
 }

dlstreamer-pipeline-server/user_scripts/gvapython/sscape/sscape_policies.py

@@ -45,8 +45,12 @@ def reidPolicy(pobj, item, fw, fh):
 
 def classificationPolicy(pobj, item, fw, fh):
   detectionPolicy(pobj, item, fw, fh)
-  # todo: add configurable parameters(set tensor name)
-  pobj['category'] = item['classification_layer_name:efficientnet-b0/model/head/dense/BiasAdd:0']['label']
+  categories = {}
+  for tensor in item.get('tensors', [{}]):
+    name = tensor.get('name','')
+    if name and name != 'detection':
+      categories[name] = tensor.get('label','')
+  pobj.update(categories)
   return
 
 def ocrPolicy(pobj, item, fw, fh):

docs/user-guide/other-topics/how-to-configure-dlstreamer-video-pipeline.md

@@ -1,6 +1,6 @@
 # How to Configure DLStreamer Video Pipeline
 
-## Video Pipeline Configuration in UI camera calibration page (in Kubernetes deployment)
+## Video Pipeline Configuration in UI Camera Calibration Page (in Kubernetes Deployment)
 
 When Intel® SceneScape is deployed in a Kubernetes environment, you can configure DLStreamer video pipelines directly through the camera calibration web interface. This provides a user-friendly way to generate and customize GStreamer pipelines for your cameras without manually editing configuration files.
 
@@ -17,17 +17,100 @@ In Kubernetes deployments, the camera calibration form provides access to a subs
 
 #### Core Pipeline Fields
 
-- **Camera (Video Source)**: specifies the video source command. Supported formats:
+- **Camera (Video Source)**: Specifies the video source command. Supported formats:
   - RTSP streams: `rtsp://camera-ip:554/stream` (raw H.264).
   - HTTP/HTTPS streams: `http://camera-ip/mjpeg` (MJPEG).
   - File sources: `file://video.ts` (relative to video folder).
-- **Camera Chain**: defines the sequence or combination of AI models to chain together in the pipeline using their short identifiers (e.g., "retail"). Models can be chained serially (one after another) or in parallel arrangements. These identifiers are defined in the model configuration file with their detailed parameters needed for pipeline generation. The model identifier may be optionally followed by `=` and an inference device identifier, e.g., `retail=GPU` will configure the pipeline to run the model inference on GPU. If the inference device is not specified, CPU is used as the default. See [DLStreamer documentation](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dl-streamer/dev_guide/gpu_device_selection.html) for GPU device selection convention.
+- **Camera Chain**: defines the sequence or combination of AI models to chain together in the pipeline using their short identifiers (e.g., "retail"). Models can be chained serially (one after another). For details on chaining syntax, available models, and usage examples, see the [Model Chaining](#model-chaining) section below.
+- **Camera Pipeline**: The generated or custom GStreamer pipeline string
 
-> **Note**: On systems with Intel GPU (either integrated or discrete), it is highly recommended to run both the decoding and the inference on GPU, so that other Intel® SceneScape services can fully benefit from available CPU cores.
+#### Model Chaining
 
-> **Note**: Currently, only limited model chaining is supported. See the limitations section below.
+Model chaining allows you to combine multiple AI models in a single pipeline to create more sophisticated video analytics workflows. For example, you can chain a person detection model with a re-identification model to first detect people in the video and then generate unique identifiers for tracking.
 
-- **Camera Pipeline**: The generated or custom GStreamer pipeline string
+##### Prerequisites
+
+By default, only a limited number of models is downloaded during helm chart installation, which limits the possibilities of model chaining. To enable the full set of models:
+
+1. Set `initModels.modelType=all` in `kubernetes/scenescape-chart/values.yaml`.
+2. Configure desired model precisions (e.g., `initModels.modelPrecisions=FP16`) in `kubernetes/scenescape-chart/values.yaml`.
+3. (Re)deploy SceneScape to download the additional models.
+
+##### Chaining Syntax
+
+- **Serial chaining**: Use the `+` operator to chain models sequentially (e.g., `retail+reid`).
+- **Device specification**: Optionally specify the inference device using `=` (e.g., `retail=GPU`). See [DLStreamer documentation](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dl-streamer/dev_guide/gpu_device_selection.html) for GPU device selection convention.
+- **Default device**: If no device is specified, CPU is used as the default.
+
+> **Note**: On systems with Intel GPU (either integrated or discrete), it is highly recommended to run both the decoding and the inference on GPU, so that other Intel® SceneScape services can fully benefit from available CPU cores. GPU inference typically provides better performance for complex models.
+
+**Example**: `retail=GPU+reid=GPU` runs person detection on GPU, then feeds the results to person re-identification also running on GPU.
+
+##### Available Models
+
+Use the following short names to refer to each model in the chain:
+
+| Category              | Full Model Name                              | Short Name  | Description                               |
+| --------------------- | -------------------------------------------- | ----------- | ----------------------------------------- |
+| **Person Detection**  | person-detection-retail-0013                 | retail      | General person detection                  |
+|                       | pedestrian-and-vehicle-detector-adas-0001    | pedveh      | Pedestrian and vehicle detection          |
+| **Person Analysis**   | person-reidentification-retail-0277          | reid        | Person re-identification                  |
+|                       | person-attributes-recognition-crossroad-0238 | personattr  | Person attributes (age, gender, clothing) |
+|                       | age-gender-recognition-retail-0013           | agegender   | Age and gender classification             |
+|                       | human-pose-estimation-0001                   | pose        | Human pose estimation                     |
+| **Vehicle Detection** | vehicle-detection-0200                       | veh0200     | Vehicle detection (newer model)           |
+|                       | vehicle-detection-0201                       | veh0201     | Vehicle detection (alternative)           |
+|                       | vehicle-detection-0202                       | veh0202     | Vehicle detection (alternative)           |
+|                       | vehicle-detection-adas-0002                  | vehadas     | ADAS vehicle detection                    |
+|                       | person-vehicle-bike-detection-2000           | pvb2000     | Multi-class detection                     |
+|                       | person-vehicle-bike-detection-2001           | pvb2001     | Multi-class detection (v2)                |
+|                       | person-vehicle-bike-detection-2002           | pvb2002     | Multi-class detection (v3)                |
+|                       | person-vehicle-bike-detection-crossroad-0078 | pvbcross78  | Crossroad detection                       |
+|                       | person-vehicle-bike-detection-crossroad-1016 | pvbcross16  | Crossroad detection (v2)                  |
+| **Vehicle Analysis**  | vehicle-attributes-recognition-barrier-0042  | vehattr     | Vehicle attributes (color, type)          |
+|                       | vehicle-license-plate-detection-barrier-0106 | platedetect | License plate detection                   |
+| **Text Analysis**     | horizontal-text-detection-0001               | textdetect  | Text detection                            |
+|                       | text-recognition-0012                        | textrec     | Text recognition                          |
+|                       | text-recognition-resnet-fc                   | textresnet  | ResNet-based text recognition             |
+
+##### Common Chaining Patterns
+
+**Person Analytics Workflows:**
+
+```
+# Basic person detection with re-identification
+retail+reid
+
+# Person detection with attributes analysis
+retail+personattr
+
+# Person detection with age/gender classification
+retail=GPU+agegender=GPU
+
+```
+
+**Vehicle Analytics Workflows:**
+
+```
+# Vehicle detection with re-identification
+veh0200=GPU+reid=GPU
+
+# Vehicle detection with attributes
+veh0200+vehattr
+
+# Vehicle detection with license plate detection
+veh0200+platedetect
+```
+
+**Multi-Class Detection:**
+
+```
+# Detect people, vehicles, and bikes
+pvb2000=GPU
+
+# Multi-class detection with re-identification
+pvb2000=GPU+reid=GPU
+```
 
 #### Advanced Configuration
 
@@ -36,7 +119,7 @@ In Kubernetes deployments, the camera calibration form provides access to a subs
 
 > **Note**: The `AUTO` setting for decode device does not assume the optimal setting in each possible case. There might be cases when the optimal configuration can be achieved by setting the decode device manually.
 
-> **Note**: The Model Config field references configuration files that define AI model parameters and processing settings. See [Model Configuration File Format](model-configuration-file-format.md) for more details.
+> **Note**: The Model Config field references configuration files that define AI model parameters and processing settings. The default configuration file `model_config.json` is auto-generated for the models downloaded by the SceneScape model installer. See [Model Configuration File Format](model-configuration-file-format.md) for more details on the file format and when/how it should be updated.
 
 #### Camera Intrinsics and Distortion
 
@@ -101,12 +184,11 @@ After generating a pipeline preview, you can make manual adjustments:
 
 ### Limitations
 
-- Multiple model chaining is not supported yet. Only a single detection model can be used as **Camera Chain**.
+- Only serial chaining of detectors with classification or re-identification models is supported in the **Camera Chain** field, where the ROI from the detection model serves as input to the classification or re-identification model in the chain. Serial chaining of two or more detectors is not supported (e.g. vehicle detector → license plate detector → OCR). Parallel inference on multiple models is not yet supported.
 - Distortion correction is temporarily disabled due to a bug in DLStreamer-Pipeline-Server.
 - Explicit frame rate and resolution configuration is not available yet.
 - Network instability and camera disconnects are not handled gracefully for network-based streams (RTSP/HTTP/HTTPS) and may cause the pipeline to fail.
 - Cross-stream batching is not supported since in Intel® SceneScape Kubernetes deployment each camera pipeline is running in a separate Pod.
-- The input format section in the model config JSON file is currently ignored. This results in GStreamer automatically finding the best possible input format for a model. If this is not sufficient, edit the pipeline string directly in the UI **Camera Pipeline** field to set arbitrary video formats.
 - Direct selection of a specific GPU as decode device on systems with multiple GPUs is not supported. As a workaround, use specific GStreamer elements in the **Camera Pipeline** field according to [DLStreamer documentation](https://docs.openedgeplatform.intel.com/dev/edge-ai-libraries/dl-streamer/dev_guide/gpu_device_selection.html).
 
 ### Troubleshooting