v3.21.0

.config

@@ -1 +1 @@
-hailort_version=4.8.1
+hailort_version=4.10.0

README.rst

@@ -54,7 +54,8 @@ Prerequisites
 * At least 6GB's of free disk space
 
 
-   Note: This version is compatible with HailoRT v4.8.1.
+.. note::
+    This version is compatible with HailoRT v4.10.0.
 
 
 Installation
@@ -78,6 +79,9 @@ Installation
    * - Yocto installation
      - `Read more about Yocto installation <docs/installation/yocto.rst>`_
      - Yocto supported BSP's
+   * - Raspberry pi installation
+     - `Read more about Raspberry pi installation <docs/installation/raspberry-pi-install.rst>`_
+     - Raspberry Pi OS, Ubuntu 20.04
 
 
 
@@ -100,9 +104,11 @@ Example Applications built with TAPPAS
 TAPPAS comes with a rich set of pre-configured pipelines optimized for different common hosts.
 
 
-   **NOTE:** Architecture-specific application examples (i.MX, Raspberry PI, etc..) use platform-specific hardware accelerators and are not compatible with different architectures.
+.. note::
+    Architecture-specific application examples (i.MX, Raspberry PI, etc..) use platform-specific hardware accelerators and are not compatible with different architectures.
 
-   **NOTE:** Running application examples requires a direct connection to a monitor.
+.. note::
+    Running application examples requires a direct connection to a monitor.
 
 
 Basic Single Network Pipelines
@@ -262,6 +268,25 @@ The pipeline demonstrates inference based decision making (Vehicle detection) fo
 Changelog
 ----------
 
+**v3.21.0 (October 2022)**
+
+* New Apps:
+
+  * `Multi-stream detection that uses HailoRT Stream Multiplexer <apps/gstreamer/general/multistream_detection/README.rst>`_ - Demonstrates the usage of HailoRT stream multiplexer (preview)
+
+* New elements - `hailoexportfile` and `hailoexportmq` which provide an access point in the pipeline to export metadata (HailoObjects)
+* Improved pipeline profiling by adding new tracers and replacing the GUI of `gst-shark <docs/write_your_own_application/debugging.rst>`_
+* Ubuntu 22 is now supported (GStreamer 1.20, preview)
+* Yocto Kirkstone is now supported (GStreamer 1.20)
+
+**v3.20.0 (August 2022)**
+
+* New Apps:
+
+  * `Detection every X frames pipeline <apps/gstreamer/general/detection/README.rst>`_ - Demonstrates the ability of skipping frames using a tracker
+
+* Improvements to Multi-Camera Multi-Person Tracking (RE-ID) pipeline (released)
+
 **v3.19.1 (July 2022)**
 
 * New Apps:
@@ -271,7 +296,7 @@ Changelog
 **v3.19.0 (June 2022)**
 
 * New Apps:
-  
+
   * Added Cascading networks, Depth Estimation, Pose Estimation and Semantic Segmentation pipelines for `i.MX Pipelines <apps/gstreamer/imx/README.rst>`_
 
 * Added an option to control post-process parameters via a JSON configuration for the detection application
@@ -280,9 +305,11 @@ Changelog
 * LPR (License Plate Recognition) pipeline is simplified to bash only
 * New detection post-process - Nanodet
 
-.. note:: Ubuntu 18.04 will be deprecated in TAPPAS future version
+.. note::
+    Ubuntu 18.04 will be deprecated in TAPPAS future version
 
-.. note:: Python 3.6 will be deprecated in TAPPAS future version
+.. note::
+    Python 3.6 will be deprecated in TAPPAS future version
 
 **v3.18.0 (April 2022)**
 

apps/gstreamer/general/README.rst

@@ -14,7 +14,7 @@ That's a great question! Hailo provides a `Sanity Pipeline <sanity_pipeline/READ
 #. `Multinetworks parallel <multinetworks_parallel/README.rst>`_ - single-stream multi-networks pipeline on top of GStreamer using the Hailo-8 device.
 #. `Instance segmentation <instance_segmentation/README.rst>`_ - single-stream instance segmentation on top of GStreamer using the Hailo-8 device.
 #. `Multi-stream detection <multistream_detection/README.rst>`_ - Multi stream object detection (up to 8 RTSP camera into one Hailo-8 chip).
-#. `Pose Estimation <pose_estimation/../README.rst>`_ - Human pose estimation using ``centerpose`` network.
+#. `Pose Estimation <pose_estimation/README.rst>`_ - Human pose estimation using ``centerpose`` network.
 #. `Segmentation <segmentation/README.rst>`_ - Semantic segmentation using ``resnet18_fcn8`` network on top of GStreamer using the Hailo-8 device.
 #. `Facial landmark <facial_landmarks/README.rst>`_ - Facial landmarking application.
 #. `Face Detection <face_detection/README.rst>`_ - Face detection application.

apps/gstreamer/general/cascading_networks/README.rst

@@ -41,9 +41,10 @@ The output should look like:
 Model
 -----
 
+Joined together:
 
-* ``lightface_slim`` in resolution of 320X240X3 - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/lightface_slim.yaml.
-* ``tddfa_mobilenet_v1`` in resolution of 120X120X3 - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/tddfa_mobilenet_v1.yaml.
+* ``lightface_slim`` in resolution of 320X240X3 - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/lightface_slim.yaml
+* ``tddfa_mobilenet_v1`` in resolution of 120X120X3 - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/tddfa_mobilenet_v1.yaml
 
 How it works?
 -------------

apps/gstreamer/general/century/README.rst

@@ -30,7 +30,7 @@ Supported Networks
 ------------------
 
 
-* 'yolox_l' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/yolox_l_leaky.yaml
+* 'yolox_l_leaky' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/yolox_l_leaky.yaml
 
 Run
 ---
@@ -54,3 +54,21 @@ How it works
 ------------
 
 This app is based on our `multi device pipeline template <../../../../docs/pipelines/multi_device.rst>`_
+
+How to use Retraining to replace models
+---------------------------------------
+
+.. note:: It is recommended to first read the :ref:`Retraining TAPPAS Models<retraining_tappas_models>` page. 
+
+You can use Retraining Dockers (available on Hailo Model Zoo), to replace the following models with ones
+that are trained on your own dataset:
+
+- ``yolox_l``
+
+  - `Retraining docker <https://github.com/hailo-ai/hailo_model_zoo/tree/master/training/yolox>`_
+
+   - For best compatibility and performance with TAPPAS, use for compilation the corresponsing YAML file from above.
+  - TAPPAS changes to replace model:
+
+    - Update HEF_PATH on the .sh file
+    - Update ``resources/configs/yolox.json`` with your new post-processing parameters (NMS)

apps/gstreamer/general/century/century.sh

@@ -28,6 +28,7 @@ function init_variables() {
     additonal_parameters=""
     sync_pipeline=false
     max_devices=$(lspci -d 1e60: | wc -l)
+    max_devices=$(($max_devices>4 ? 4 : $max_devices))
 
     if (($max_devices == 0)); then
         echo "Error: No devices found."
@@ -65,7 +66,7 @@ function parse_args() {
             print_gst_launch_only=true
         elif [ "$1" = "--show-fps" ]; then
             echo "Printing fps"
-            additonal_parameters="-v 2>&1 | grep -e hailo_display -e hailodevicestats"
+            additonal_parameters="-v | grep -e hailo_display -e hailodevicestats"
         elif [ "$1" = "--input" ] || [ "$1" == "-i" ]; then
             input_source="$2"
             shift

apps/gstreamer/general/classification/classification.sh

@@ -48,7 +48,7 @@ function parse_args() {
     while test $# -gt 0; do
         if [ "$1" = "--show-fps" ]; then
             echo "Printing fps"
-            additonal_parameters="-v 2>&1 | grep hailo_display"
+            additonal_parameters="-v | grep hailo_display"
         elif [ "$1" = "--print-gst-launch" ]; then
             print_gst_launch_only=true
         elif [ "$1" = "--input" ] || [ "$1" == "-i" ]; then

apps/gstreamer/general/depth_estimation/README.rst

@@ -42,7 +42,7 @@ Model
 -----
 
 
-* ``fast_depth`` in resolution of 224X224X3.
+* ``fast_depth`` in resolution of 224X224X3: https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/fast_depth.yaml
 
 How it works
 ------------

apps/gstreamer/general/depth_estimation/depth_estimation.sh

@@ -44,7 +44,7 @@ function parse_args() {
     while test $# -gt 0; do
         if [ "$1" = "--show-fps" ]; then
             echo "Printing fps"
-            additonal_parameters="-v 2>&1 | grep hailo_display"
+            additonal_parameters="-v | grep hailo_display"
         elif [ "$1" = "--print-gst-launch" ]; then
             print_gst_launch_only=true
         elif [ "$1" = "--input" ] || [ "$1" == "-i" ]; then

apps/gstreamer/general/detection/README.rst

@@ -32,10 +32,11 @@ Supported Networks
 ------------------
 
 
-* 'yolov5' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/yolov5m.yaml
-* 'yolov4' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/yolov4_leaky.yaml
-* 'yolov3' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/yolov3_gluon.yaml
+* 'yolov5m_wo_spp_60p' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/yolov5m_wo_spp_60p.yaml
+* 'yolov4_leaky' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/yolov4_leaky.yaml
+* 'yolov3_gluon' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/yolov3_gluon.yaml
 * 'mobilenet_ssd' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/ssd_mobilenet_v1.yaml
+* 'nanodet_repvgg' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/nanodet_repvgg.yaml
 
 Run
 ---
@@ -57,3 +58,59 @@ How it works?
 -------------
 
 This app is based on our `single network pipeline template <../../../../docs/pipelines/single_network.rst>`_
+
+How to use Retraining to replace models
+---------------------------------------
+
+.. note:: It is recommended to first read the :ref:`Retraining TAPPAS Models<retraining_tappas_models>` page. 
+
+You can use Retraining Dockers (available on Hailo Model Zoo), to replace the following models with ones
+that are trained on your own dataset:
+
+- ``yolov5m``
+
+  - `Retraining docker <https://github.com/hailo-ai/hailo_model_zoo/tree/master/training/yolov5>`_
+
+    - For best compatibility and performance with TAPPAS, use for compilation the corresponsing YAML file from above.
+  - TAPPAS changes to replace model:
+
+    - Update HEF_PATH on the .sh file
+    - Update ``resources/configs/yolov5.json`` with your new post-processing parameters (NMS)
+- ``yolov4``
+
+  - `Retraining docker <https://github.com/hailo-ai/hailo_model_zoo/tree/master/training/yolov4>`_
+
+    - For best compatibility and performance with TAPPAS, use for compilation the corresponsing YAML file from above.
+  - TAPPAS changes to replace model:
+
+    - Update HEF_PATH on the .sh file
+    - Update ``resources/configs/yolov4.json`` with your new post-processing parameters (NMS)
+- ``yolov3``
+
+  - `Retraining docker <https://github.com/hailo-ai/hailo_model_zoo/tree/master/training/yolov3>`_
+
+    - For best compatibility and performance with TAPPAS, use for compilation the corresponsing YAML file from above.
+  - TAPPAS changes to replace model:
+
+    - Update HEF_PATH on the .sh file
+    - Update ``resources/configs/yolov3.json`` with your new post-processing parameters (NMS)
+- ``mobilenet_ssd``
+
+  - `Retraining docker <https://github.com/hailo-ai/hailo_model_zoo/tree/master/training/ssd>`_
+
+    - For best compatibility and performance with TAPPAS, use for compilation the corresponsing YAML file from above.
+  - TAPPAS changes to replace model:
+
+    - Update HEF_PATH on the .sh file
+    - Update `mobilenet_ssd.cpp <https://github.com/hailo-ai/tappas/blob/master/core/hailo/gstreamer/libs/postprocesses/detection/mobilenet_ssd.cpp#L141>`_
+      with your new paremeters, then recompile to create ``libmobilenet_ssd_post.so``
+- ``nanodet_repvgg``
+
+  - `Retraining docker <https://github.com/hailo-ai/hailo_model_zoo/tree/master/training/nanodet>`_
+
+    - For best compatibility and performance with TAPPAS, use for compilation the corresponsing YAML file from above.
+  - TAPPAS changes to replace model:
+
+    - Update HEF_PATH on the .sh file
+    - Update `nanodet.cpp <https://github.com/hailo-ai/tappas/blob/master/core/hailo/gstreamer/libs/postprocesses/detection/nanodet.cpp#L221>`_
+      with your new paremeters, then recompile to create ``libnanodet_post.so``

apps/gstreamer/general/detection/detection.sh

@@ -95,7 +95,7 @@ function parse_args() {
             debug_stats_export="GST_DEBUG=hailodevicestats:5"
         elif [ "$1" = "--show-fps" ]; then
             echo "Printing fps"
-            additonal_parameters="-v 2>&1 | grep -e hailo_display -e hailodevicestats"
+            additonal_parameters="-v | grep -e hailo_display -e hailodevicestats"
         elif [ "$1" = "--input" ] || [ "$1" == "-i" ]; then
             input_source="$2"
             shift