Release OpenEB 3.1.1

prophesee-ai · Nov 25, 2022 · 474368f · 474368f
1 parent dcf571e
commit 474368f
Show file tree

Hide file tree

Showing 40 changed files with 558 additions and 437 deletions.
diff --git a/.github/workflows/all_jobs.yaml b/.github/workflows/all_jobs.yaml
@@ -24,7 +24,7 @@ jobs:
       - name: Install dependencies
         run: |
           sudo apt-get -y update && sudo apt-get install -y sudo apt-utils wget cmake build-essential \
-          libglew-dev git g++ libusb-1.0-0-dev libeigen3-dev libgtest-dev libssl-dev \
+          libglew-dev git g++ libusb-1.0-0-dev libgtest-dev libssl-dev \
           python3-pip python3-distutils xvfb libglfw3 libglfw3-dev libglew-dev libopencv-dev ffmpeg \
           libboost-dev libboost-program-options-dev libboost-filesystem-dev libboost-timer-dev \
           libboost-chrono-dev libboost-thread-dev zip unzip git-lfs
@@ -71,7 +71,7 @@ jobs:
       - name: Install dependencies
         run: |
           sudo apt-get -y update && sudo apt-get install -y sudo apt-utils wget cmake build-essential \
-          libglew-dev git g++ libusb-1.0-0-dev libeigen3-dev libgtest-dev libssl-dev \
+          libglew-dev git g++ libusb-1.0-0-dev libgtest-dev libssl-dev \
           python3-pip python3-distutils xvfb libglfw3 libglfw3-dev libglew-dev libopencv-dev ffmpeg \
           libboost-dev libboost-program-options-dev libboost-filesystem-dev libboost-timer-dev \
           libboost-chrono-dev libboost-thread-dev zip unzip

diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -9,7 +9,7 @@
 
 cmake_minimum_required(VERSION 3.5)
 
-project(metavision VERSION 3.1.0)
+project(metavision VERSION 3.1.1)
 set(PROJECT_VERSION_SUFFIX "")
 
 if(PROJECT_VERSION_SUFFIX STREQUAL "")

diff --git a/README.md b/README.md
@@ -59,20 +59,22 @@ python3 -m pip install pip --upgrade
 
 To use Machine Learning features, you need to install some additional dependencies.
 
-First, if you have some Nvidia hardware with GPUs, install `CUDA (10.2, 11.1 or 11.3) <https://developer.nvidia.com/cuda-downloads>`_
+First, if you have some Nvidia hardware with GPUs, install `CUDA (10.2 or 11.1) <https://developer.nvidia.com/cuda-downloads>`_
 and `cuDNN <https://docs.nvidia.com/deeplearning/cudnn/install-guide/index.html>`_ to leverage them with pytorch and libtorch.
 
 Make sure that you install a version of CUDA that is compatible with your GPUs by checking
 `Nvidia compatibility page <https://docs.nvidia.com/deeplearning/cudnn/support-matrix/index.html>`_.
 
 Note that, at the moment, we don't support `OpenCL <https://www.khronos.org/opencl/>`_ and AMD GPUs.
 
-Then, install pytorch. Go to `pytorch.org <https://pytorch.org>`_ to retrieve the pip command that you
-will launch in a console to install PyTorch 1.8.2 LTS. Here is an example of a command that can be retrieved for
+Then, install PyTorch 1.8.2 LTS. This version was deprecated by PyTorch team but can still be downloaded
+in `the Previous Versions page of pytorch.org <https://pytorch.org/get-started/previous-versions/#v182-with-lts-support>`_
+(in future releases of Metavision ML, more recent version of PyTorch will be leveraged).
+Retrieve and execute the pip command for the installation. Here is an example of a command that can be retrieved for
 pytorch using CUDA 11.1:
 
 ```bash
-python3 -m pip install torch==1.8.2+cu111 torchvision==0.9.2+cu111 torchaudio==0.8.2 -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html
+python3 -m pip install torch==1.8.2 torchvision==0.9.2 torchaudio==0.8.2 --extra-index-url https://download.pytorch.org/whl/lts/1.8/cu111
 ```
 
 Then install some extra Python libraries:
@@ -130,17 +132,30 @@ with the following command: `sudo cmake --build . --target install`. In that cas
 `LD_LIBRARY_PATH` with `export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib` (If you want to update this path
 permanently, you should add the previous command in your ~/.bashrc)
 
-*Note* that since OpenEB 3.0.0, Prophesee camera plugins are included in the OpenEB repository, so you don't need to perform
-any extra step to install them. If you are using a third-party camera, you need to install the plugin provided
-by the camera vendor and specify the location of the plugin using the `MV_HAL_PLUGIN_PATH` environment variable.
+You can also deploy the OpenEB files (applications, samples, libraries etc.) in a directory of your choice by using 
+the `CMAKE_INSTALL_PREFIX` variable (`-DCMAKE_INSTALL_PREFIX=<OPENEB_INSTALL_DIR>`) when generating the makefiles
+in step 3. Similarly, you can configure the directory where the Python packages will be deployed using the
+`PYTHON3_SITE_PACKAGES` variable (`-DPYTHON3_SITE_PACKAGES=<PYTHON3_PACKAGES_INSTALL_DIR>`).
+
+Since OpenEB 3.0.0, Prophesee camera plugins are included in OpenEB. If you did not perform the optional deployment step
+(`sudo cmake --build . --target install`) and instead used “setup_env.sh”, then you need to copy the udev rules files 
+used by Prophesee cameras in the system path and reload them so that your camera is detected with this command:
+
+```bash
+sudo cp $METAVISION_SRC_DIR/hal_psee_plugins/resources/rules/*.rules /etc/udev/rules.d
+udevadm control --reload-rules
+udevadm trigger
+```
+
+If you are using a third-party camera, you need to install the plugin provided by the camera vendor and specify
+the location of the plugin using the `MV_HAL_PLUGIN_PATH` environment variable.
 
 To get started with OpenEB, you can download some [sample recordings](https://docs.prophesee.ai/stable/datasets.html) 
 and visualize them with [metavision_viewer](https://docs.prophesee.ai/stable/metavision_sdk/modules/driver/guides/viewer.html#chapter-sdk-driver-samples-viewer)
 or you can stream data from your Prophesee-compatible event-based camera.
 
 ### Running the test suite (Optional)
 
-
 Running the test suite is a sure-fire way to ensure you did everything well with your compilation and installation process.
 
  * Download [the files](https://dataset.prophesee.ai/index.php/s/hyCzGM4tpR8w5bx) necessary to run the tests.
@@ -232,7 +247,7 @@ python -m pip install pip --upgrade
 
 To use Machine Learning features, you need to install some additional dependencies.
 
-First, if you have some Nvidia hardware with GPUs, install `CUDA (10.2, 11.1 or 11.3) <https://developer.nvidia.com/cuda-downloads>`_
+First, if you have some Nvidia hardware with GPUs, install `CUDA (10.2 or 11.1) <https://developer.nvidia.com/cuda-downloads>`_
 and `cuDNN <https://docs.nvidia.com/deeplearning/cudnn/install-guide/index.html>`_ to leverage them with pytorch and libtorch.
 
 Then, install pytorch. Go to `pytorch.org <https://pytorch.org>`_ to retrieve the pip command that you

diff --git a/cmake/custom_functions/add_library_version_header.cmake b/cmake/custom_functions/add_library_version_header.cmake
@@ -7,9 +7,9 @@
 # on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and limitations under the License.
 
-set(GIT_BRANCH "main")
-set(GIT_COMMIT_ID "4c19f90e13c20548513c88b89c4e58bc8de8a1bf")
-set(GIT_COMMIT_DATE "2022-10-13 16:56:39 +0200")
+set(GIT_BRANCH "HEAD")
+set(GIT_COMMIT_ID "37aa7510f3d6bdbd2b2b66e816d4fb0e3a420e9d")
+set(GIT_COMMIT_DATE "2022-11-23 15:21:14 +0100")
 
 # If git information are not provided in command line when running cmake, try to automatically determine them
 if(NOT GIT_BRANCH OR NOT GIT_COMMIT_ID OR NOT GIT_COMMIT_DATE)

diff --git a/cmake/custom_targets/README_metavision_open.md b/cmake/custom_targets/README_metavision_open.md
@@ -59,20 +59,22 @@ python3 -m pip install pip --upgrade
 
 To use Machine Learning features, you need to install some additional dependencies.
 
-First, if you have some Nvidia hardware with GPUs, install `CUDA (10.2, 11.1 or 11.3) <https://developer.nvidia.com/cuda-downloads>`_
+First, if you have some Nvidia hardware with GPUs, install `CUDA (10.2 or 11.1) <https://developer.nvidia.com/cuda-downloads>`_
 and `cuDNN <https://docs.nvidia.com/deeplearning/cudnn/install-guide/index.html>`_ to leverage them with pytorch and libtorch.
 
 Make sure that you install a version of CUDA that is compatible with your GPUs by checking
 `Nvidia compatibility page <https://docs.nvidia.com/deeplearning/cudnn/support-matrix/index.html>`_.
 
 Note that, at the moment, we don't support `OpenCL <https://www.khronos.org/opencl/>`_ and AMD GPUs.
 
-Then, install pytorch. Go to `pytorch.org <https://pytorch.org>`_ to retrieve the pip command that you
-will launch in a console to install PyTorch 1.8.2 LTS. Here is an example of a command that can be retrieved for
+Then, install PyTorch 1.8.2 LTS. This version was deprecated by PyTorch team but can still be downloaded
+in `the Previous Versions page of pytorch.org <https://pytorch.org/get-started/previous-versions/#v182-with-lts-support>`_
+(in future releases of Metavision ML, more recent version of PyTorch will be leveraged).
+Retrieve and execute the pip command for the installation. Here is an example of a command that can be retrieved for
 pytorch using CUDA 11.1:
 
 ```bash
-python3 -m pip install torch==1.8.2+cu111 torchvision==0.9.2+cu111 torchaudio==0.8.2 -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html
+python3 -m pip install torch==1.8.2 torchvision==0.9.2 torchaudio==0.8.2 --extra-index-url https://download.pytorch.org/whl/lts/1.8/cu111
 ```
 
 Then install some extra Python libraries:
@@ -130,17 +132,30 @@ with the following command: `sudo cmake --build . --target install`. In that cas
 `LD_LIBRARY_PATH` with `export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib` (If you want to update this path
 permanently, you should add the previous command in your ~/.bashrc)
 
-*Note* that since OpenEB 3.0.0, Prophesee camera plugins are included in the OpenEB repository, so you don't need to perform
-any extra step to install them. If you are using a third-party camera, you need to install the plugin provided
-by the camera vendor and specify the location of the plugin using the `MV_HAL_PLUGIN_PATH` environment variable.
+You can also deploy the OpenEB files (applications, samples, libraries etc.) in a directory of your choice by using 
+the `CMAKE_INSTALL_PREFIX` variable (`-DCMAKE_INSTALL_PREFIX=<OPENEB_INSTALL_DIR>`) when generating the makefiles
+in step 3. Similarly, you can configure the directory where the Python packages will be deployed using the
+`PYTHON3_SITE_PACKAGES` variable (`-DPYTHON3_SITE_PACKAGES=<PYTHON3_PACKAGES_INSTALL_DIR>`).
+
+Since OpenEB 3.0.0, Prophesee camera plugins are included in OpenEB. If you did not perform the optional deployment step
+(`sudo cmake --build . --target install`) and instead used “setup_env.sh”, then you need to copy the udev rules files 
+used by Prophesee cameras in the system path and reload them so that your camera is detected with this command:
+
+```bash
+sudo cp $METAVISION_SRC_DIR/hal_psee_plugins/resources/rules/*.rules /etc/udev/rules.d
+udevadm control --reload-rules
+udevadm trigger
+```
+
+If you are using a third-party camera, you need to install the plugin provided by the camera vendor and specify
+the location of the plugin using the `MV_HAL_PLUGIN_PATH` environment variable.
 
 To get started with OpenEB, you can download some [sample recordings](https://docs.prophesee.ai/stable/datasets.html) 
 and visualize them with [metavision_viewer](https://docs.prophesee.ai/stable/metavision_sdk/modules/driver/guides/viewer.html#chapter-sdk-driver-samples-viewer)
 or you can stream data from your Prophesee-compatible event-based camera.
 
 ### Running the test suite (Optional)
 
-
 Running the test suite is a sure-fire way to ensure you did everything well with your compilation and installation process.
 
  * Download [the files](https://dataset.prophesee.ai/index.php/s/hyCzGM4tpR8w5bx) necessary to run the tests.
@@ -232,7 +247,7 @@ python -m pip install pip --upgrade
 
 To use Machine Learning features, you need to install some additional dependencies.
 
-First, if you have some Nvidia hardware with GPUs, install `CUDA (10.2, 11.1 or 11.3) <https://developer.nvidia.com/cuda-downloads>`_
+First, if you have some Nvidia hardware with GPUs, install `CUDA (10.2 or 11.1) <https://developer.nvidia.com/cuda-downloads>`_
 and `cuDNN <https://docs.nvidia.com/deeplearning/cudnn/install-guide/index.html>`_ to leverage them with pytorch and libtorch.
 
 Then, install pytorch. Go to `pytorch.org <https://pytorch.org>`_ to retrieve the pip command that you

diff --git a/cmake/custom_targets/create_metavision_open_archive_folder.cmake b/cmake/custom_targets/create_metavision_open_archive_folder.cmake
@@ -48,6 +48,10 @@ file(COPY "${PROJECT_SOURCE_DIR}/utils/github_actions/openeb/"
 file(REMOVE_RECURSE "${OUTPUT_DIR}/cmake/custom_targets_metavision_sdk")
 file(REMOVE "${OUTPUT_DIR}/cmake/custom_functions/create_addon_module_archive_folder.cmake")
 file(REMOVE "${OUTPUT_DIR}/cmake/custom_functions/documentation.cmake")
+# Remove unwanted plugins
+file(REMOVE "${OUTPUT_DIR}/hal_psee_plugins/src/plugin/psee_gen4_evk1.cpp")
+file(REMOVE "${OUTPUT_DIR}/hal_psee_plugins/src/plugin/psee_gen4_evk2.cpp")
+file(REMOVE_RECURSE "${OUTPUT_DIR}/hal_psee_plugins/src/plugin/raven")
 # Remove unwanted files
 file(REMOVE "${OUTPUT_DIR}/sdk/cmake/MetavisionSDKCPackConfig.cmake")
 file(REMOVE "${OUTPUT_DIR}/sdk/cmake/MetavisionStudioCPackConfig.cmake")

diff --git a/hal/cpp/include/metavision/hal/facilities/i_ll_biases.h b/hal/cpp/include/metavision/hal/facilities/i_ll_biases.h
@@ -40,4 +40,4 @@ class I_LL_Biases : public I_RegistrableFacility<I_LL_Biases> {
 
 } // namespace Metavision
 
-#endif // METAVISION_HAL_I_LL_BIASES_H
+#endif // METAVISION_HAL_I_LL_BIASES_H
diff --git a/hal/cpp/samples/metavision_hal_sync/metavision_hal_sync.cpp b/hal/cpp/samples/metavision_hal_sync/metavision_hal_sync.cpp
@@ -230,7 +230,7 @@ int main(int argc, char *argv[]) {
     const int fps       = 25; // event-based cameras do not have a frame rate, but we need one for visualization
     const int wait_time = static_cast<int>(std::round(1.f / fps * 1000)); // how much we should wait between two frames
     cv::Mat display;                                                      // frame where events will be accumulated
-    const std::string window_name = "Metavision HAL Sync";
+    const std::string window_name = (mode_master) ? "Metavision HAL Sync - Master " : "Metavision HAL Sync - Slave ";
     cv::namedWindow(window_name, cv::WINDOW_GUI_EXPANDED);
     cv::resizeWindow(window_name, i_geometry->get_width(), i_geometry->get_height());
 

diff --git a/hal_psee_plugins/CMakeLists.txt b/hal_psee_plugins/CMakeLists.txt
@@ -11,6 +11,10 @@ if(EXISTS "${CMAKE_CURRENT_LIST_DIR}/src/devices/gen4")
     set(HAL_GEN4_SUPPORT TRUE)
 endif(EXISTS "${CMAKE_CURRENT_LIST_DIR}/src/devices/gen4")
 
+if(EXISTS "${CMAKE_CURRENT_LIST_DIR}/src/devices/raven")
+    set(HAL_RAVEN_SUPPORT TRUE)
+endif(EXISTS "${CMAKE_CURRENT_LIST_DIR}/src/devices/raven")
+
 # Add libraries (plugins)
 set(biasgen_calib_files_dir "${GENERATE_FILES_DIRECTORY}/biasgen_calib")
 set(metavision_hal_psee_plugins_include_dir "${CMAKE_CURRENT_SOURCE_DIR}/include")

diff --git a/hal_psee_plugins/include/boards/treuzell/tz_libusb.h b/hal_psee_plugins/include/boards/treuzell/tz_libusb.h
@@ -12,8 +12,12 @@
 #ifndef METAVISION_HAL_TZ_LIBUSB_H
 #define METAVISION_HAL_TZ_LIBUSB_H
 
+#ifdef _MSC_VER
+#define NOMINMAX // libusb.h includes windows.h which defines min max macros that we don't want
+#endif
+
 #include <system_error>
-#include <libusb.h>
+#include <libusb-1.0/libusb.h>
 
 namespace Metavision {
 

diff --git a/...see_plugins/src/plugin/psee_gen4_evk1.cpp → ...gins/include/devices/imx636/imx636_bias.h b/...see_plugins/src/plugin/psee_gen4_evk1.cpp → ...gins/include/devices/imx636/imx636_bias.h
@@ -9,28 +9,45 @@
  * See the License for the specific language governing permissions and limitations under the License.                 *
  **********************************************************************************************************************/
 
-#if !defined(__ANDROID__) || defined(ANDROID_USES_LIBUSB)
-#include "boards/fx3/fx3_camera_discovery.h"
-#include "devices/gen4/gen4_evk1_facilities_builder.h"
-#endif
-#include "boards/rawfile/psee_file_discovery.h"
-#include "devices/utils/device_system_id.h"
-#include "metavision/hal/plugin/plugin.h"
-#include "metavision/hal/plugin/plugin_entrypoint.h"
-#include "metavision/hal/utils/hal_software_info.h"
-#include "plugin/psee_plugin.h"
-
-void initialize_plugin(void *plugin_ptr) {
-    using namespace Metavision;
-
-    Plugin &plugin = plugin_cast(plugin_ptr);
-    initialize_psee_plugin(plugin);
-
-#if !defined(__ANDROID__) || defined(ANDROID_USES_LIBUSB)
-    // Register live camera discoveries
-    auto &fx3_disc = plugin.add_camera_discovery(std::make_unique<Fx3CameraDiscovery>());
-    fx3_disc.register_device_builder(SYSTEM_CCAM3_GEN4, build_gen4_evk1_device);
-#endif
-
-    auto &file_disc = plugin.add_file_discovery(std::make_unique<PseeFileDiscovery>());
-}
+#ifndef METAVISION_HAL_IMX636_BIASES_H
+#define METAVISION_HAL_IMX636_BIASES_H
+
+#include <string>
+#include <map>
+
+#include "metavision/hal/facilities/i_ll_biases.h"
+
+namespace Metavision {
+
+static constexpr uint32_t BIAS_CONF = 0x11A10000;
+
+class Imx636LLBias {
+public:
+    Imx636LLBias(bool modifiable, std::string register_name, int sensor_offset, int current_value, int factory_default,
+                 int min_offset, int max_offset);
+    ~Imx636LLBias();
+    const std::string &get_register_name() const;
+    bool is_modifiable() const;
+    int get_min_offset();
+    int get_max_offset();
+    int get_current_offset();
+    void set_current_offset(const int val);
+    int get_current_value();
+    void set_current_value(const int val);
+    int get_factory_default_value();
+    void set_factory_default_value(const int val);
+    void display_bias();
+
+private:
+    std::string register_name_;
+    bool modifiable_;
+    int current_value_;
+    int current_offset_;
+    int factory_default_;
+    int min_offset_;
+    int max_offset_;
+};
+
+} // namespace Metavision
+
+#endif // METAVISION_HAL_IMX636_BIASES_H
diff --git a/hal_psee_plugins/include/devices/imx636/imx636_evk2_issd.h b/hal_psee_plugins/include/devices/imx636/imx636_evk2_issd.h
@@ -31,7 +31,7 @@
 
 namespace Metavision {
 
-const std::vector<RegisterOperation> issd_evk2_imx636_init = {
+static const std::vector<RegisterOperation> issd_evk2_imx636_init = {
     RegisterOperation::Write(0x00000004, 0x00000111),
     RegisterOperation::Write(0x00000000, 0x00000089),
     RegisterOperation::Write(0x00000000, 0x0000008D),
@@ -138,7 +138,7 @@ const std::vector<RegisterOperation> issd_evk2_imx636_init = {
     RegisterOperation::Write(0x00109000, 0x00000200)
 };
 
-const std::vector<RegisterOperation> issd_evk2_imx636_start = {
+static const std::vector<RegisterOperation> issd_evk2_imx636_start = {
     RegisterOperation::Write(0x00000044, 0x00000001),
     RegisterOperation::Write(0x00000014, 0x00000001),
     RegisterOperation::Write(0x0070F000, 0x00400001),
@@ -152,7 +152,7 @@ const std::vector<RegisterOperation> issd_evk2_imx636_start = {
     RegisterOperation::WriteField(0x00000008, 0x0000004F, 0x1)
 };
 
-const std::vector<RegisterOperation> issd_evk2_imx636_stop = {
+static const std::vector<RegisterOperation> issd_evk2_imx636_stop = {
     RegisterOperation::Write(0x00000018, 0x00000000),
     RegisterOperation::WriteField(0x00000008, 0x0000004E, 0x00000001),
     // Analog STOP
@@ -166,7 +166,7 @@ const std::vector<RegisterOperation> issd_evk2_imx636_stop = {
     RegisterOperation::Delay(300)
 };
 
-const std::vector<RegisterOperation> issd_evk2_imx636_destroy = {
+static const std::vector<RegisterOperation> issd_evk2_imx636_destroy = {
     // Analog DESTROY
     RegisterOperation::Write(0x00100070, 0x00400008),
     RegisterOperation::Write(0x0010006C, 0x0EE47114),
@@ -207,7 +207,7 @@ const std::vector<RegisterOperation> issd_evk2_imx636_destroy = {
     RegisterOperation::Write(0x00000004, 0x00000000)
 };
 
-Issd issd_evk2_imx636_sequence = Issd{
+static Issd issd_evk2_imx636_sequence = Issd{
     issd_evk2_imx636_init,
     issd_evk2_imx636_start,
     issd_evk2_imx636_stop,
Original file line number	Diff line number	Diff line change
Expand Up		@@ -40,4 +40,4 @@ class I_LL_Biases : public I_RegistrableFacility<I_LL_Biases> {

		} // namespace Metavision

		#endif // METAVISION_HAL_I_LL_BIASES_H
		#endif // METAVISION_HAL_I_LL_BIASES_H