Merge pull request #30 from antoniomtz/loss-prevention

Loss prevention

docs_src/use-cases/loss-prevention/advanced.md

@@ -0,0 +1,36 @@
+# Advanced Settings
+
+To further customize a loss prevention pipeline, let's add more variables to the execution:
+
+!!! Example
+
+    ```bash
+    make PIPELINE_SCRIPT=yolov8s_roi.sh RESULTS_DIR="../render_results"  run-render-mode
+    ```
+
+The above command will execute a DLStreamer pipeline using YOLOv8s model for object detection on a region of interest (ROI) with object tracking mechanism.
+
+## Modify ROI coordinates
+
+To modify the ROI coordinates, locate the file `roi.json` under src/pipelines/roi.json. Since the "objects" attribute is an array, it is possible to add multiple ROIs.
+
+```json
+[
+    {
+        "objects": [
+            {
+                "detection": {
+                    "label": "ROI1"
+                },
+                "x": 0,
+                "y": 0,
+                "w": 620,
+                "h": 1080
+            }           
+        ]
+    }
+]
+```
+For enviroments variables, follow the same tutorial as the automated self checkout [HERE](../automated-self-checkout/advanced.md)
+
+

docs_src/use-cases/loss-prevention/getting_started.md

@@ -0,0 +1,158 @@
+# Getting Started
+
+## Step by step instructions:
+
+1. Download the models using download_models/downloadModels.sh
+
+    ```bash
+    make download-models
+    ```
+
+2. Update github submodules
+
+    ```bash
+    make update-submodules
+    ```
+
+3. Download sample videos used by the performance tools
+
+    ```bash
+    make download-sample-videos
+    ```
+
+4. Build the demo Docker image
+
+    ```bash
+    make build
+    ```
+
+5. Start Loss prevention using the Docker Compose file. The Docker Compose also includes an RTSP camera simulator that will infinitely loop through the sample videos downloaded in step 3.
+
+    ```bash
+    make run-render-mode
+    ```
+
+6. Verify Docker containers
+
+    ```bash
+    docker ps --format 'table{{.Names}}\t{{.Status}}\t{{.Image}}'
+    ```
+    Result:
+    ```bash
+    NAMES                 STATUS          IMAGE
+    camera-simulator0    Up 17 seconds   jrottenberg/ffmpeg:4.1-alpine
+    src-OvmsClientGst-1  Up 17 seconds   dlstreamer:dev
+    camera-simulator     Up 17 seconds   aler9/rtsp-simple-server
+    ```
+
+7. Verify Results
+
+    After starting Automated Self Checkout you will begin to see result files being written into the results/ directory. Here are example outputs from the 3 log files.
+
+    gst-launch_<time>_gst.log
+    ```
+    /GstPipeline:pipeline0/GstGvaWatermark:gvawatermark0/GstCapsFilter:capsfilter1: caps = video/x-raw(memory:VASurface), format=(string)RGBA
+    /GstPipeline:pipeline0/GstFPSDisplaySink:fpsdisplaysink0/GstXImageSink:ximagesink0: sync = true
+    Got context from element 'vaapipostproc1': gst.vaapi.Display=context, gst.vaapi.Display=(GstVaapiDisplay)"\(GstVaapiDisplayGLX\)\ vaapidisplayglx0", gst.vaapi.Display.GObject=(GstObject)"\(GstVaapiDisplayGLX\)\ vaapidisplayglx0";
+    Progress: (open) Opening Stream
+    Pipeline is PREROLLED ...
+    Prerolled, waiting for progress to finish...
+    Progress: (connect) Connecting to rtsp://localhost:8554/camera_0
+    Progress: (open) Retrieving server options
+    Progress: (open) Retrieving media info
+    Progress: (request) SETUP stream 0
+    ```
+
+    pipeline<time>_gst.log
+    ```
+    14.58
+    14.58
+    15.47
+    15.47
+    15.10
+    15.10
+    14.60
+    14.60
+    14.88
+    14.88
+    ```
+
+    r<time>_gst.jsonl
+```json
+
+{
+    "objects": [
+        {
+            "detection": {
+                "bounding_box": {
+                    "x_max": 0.7873924346958825,
+                    "x_min": 0.6701603093745723,
+                    "y_max": 0.7915918938548927,
+                    "y_min": 0.14599881349270305
+                },
+                "confidence": 0.8677337765693665,
+                "label": "bottle",
+                "label_id": 39
+            },
+            "h": 697,
+            "region_id": 610,
+            "roi_type": "bottle",
+            "w": 225,
+            "x": 1287,
+            "y": 158
+        },
+        {
+            "detection": {
+                "bounding_box": {
+                    "x_max": 0.3221945836811382,
+                    "x_min": 0.19950163649114616,
+                    "y_max": 0.7924592239981934,
+                    "y_min": 0.1336837231479251
+                },
+                "confidence": 0.8625879287719727,
+                "label": "bottle",
+                "label_id": 39
+            },
+            "h": 711,
+            "region_id": 611,
+            "roi_type": "bottle",
+            "w": 236,
+            "x": 383,
+            "y": 144
+        },
+        {
+            "detection": {
+                "bounding_box": {
+                    "x_max": 0.5730873789069046,
+                    "x_min": 0.42000878963365595,
+                    "y_max": 0.9749763191740435,
+                    "y_min": 0.12431765065780453
+                },
+                "confidence": 0.854443371295929,
+                "label": "bottle",
+                "label_id": 39
+            },
+            "h": 919,
+            "region_id": 612,
+            "roi_type": "bottle",
+            "w": 294,
+            "x": 806,
+            "y": 134
+        }
+    ],
+    "resolution": {
+        "height": 1080,
+        "width": 1920
+    },
+    "timestamp": 755106652
+}
+
+```
+
+8. Stop the demo using docker compose down
+```bash
+make down
+```
+
+
+## [Proceed to Advanced Settings](advanced.md)    

docs_src/use-cases/loss-prevention/loss-prevention.md

@@ -0,0 +1,13 @@
+# Intel® Loss Prevention Reference Package
+
+## Overview
+
+As computer vision technology becomes more mainstream in industrial and retail settings, using it for loss prevention is becoming increasingly complex. These vision workloads are substantial and require multiple stages of processing. For example, a typical loss prevention pipeline might capture video data, define regions of interest, implement tracking to monitor which products customers interact with, analyze the data using models like YOLOv5 and EfficientNet, and then post-process it to generate metadata that highlights which products are being purchased or potentially stolen. This is just one example of how such models and workflows can be utilized.
+
+Implementing loss prevention solutions in retail isn't straightforward. Retailers, independent software vendors (ISVs), and system integrators (SIs) need a solid understanding of both hardware and software, as well as the costs involved in setting up and scaling these systems. Given the data-intensive nature of vision workloads, systems must be carefully designed, built, and deployed with numerous considerations in mind. Effectively combating shrinkage requires the right mix of hardware, software, and optimized configurations.
+
+The Intel® Loss Prevention Reference Package is designed to help with this. It provides the essential components needed to develop and deploy a loss prevention solution using Intel® hardware, software, and open-source tools. This reference implementation includes a pre-configured pipeline that's optimized for Intel® hardware, simplifying the setup of an effective computer vision-based loss prevention system for retailers.
+
+## Next Steps
+
+To begin using the loss prevention solution you can follow the [Getting Started Guide](./getting_started.md). 

mkdocs.yml

@@ -56,6 +56,10 @@ nav:
       - AI Connect for Scientific Data (AiCSD):
           - Overview: 'use-cases/AiCSD/aicsd.md'
           - GRPC Yolov5s Pipeline: 'use-cases/AiCSD/pipeline-grpc-go.md'
+      - Loss Prevention:
+          - Overview: 'use-cases/loss-prevention/loss-prevention.md'
+          - Getting Started: 'use-cases/loss-prevention/getting_started.md'
+          - Advanced Settings: 'use-cases/loss-prevention/advanced.md'
   - Releases: 'releasenotes.md'
   - Troubleshooting: 'troubleshooting.md'
 extra_css: