Tx2 install dev (#384)

* Installation script for TX2.
Updated installation.md and changed import statements in object_detection_2d gem and detr.

* Update installation.md

* Updated tx2 installation script

Tested it twice on two different TX2 boards successfully.

* Update install_tx2.sh

* Update installation.md

* Update installation.md

* Update installation.md

* Update docs/reference/installation.md

Co-authored-by: ad-daniel <44834743+ad-daniel@users.noreply.github.com>

* Update bin/install_tx2.sh

Co-authored-by: ad-daniel <44834743+ad-daniel@users.noreply.github.com>

* Update bin/install_tx2.sh

Co-authored-by: ad-daniel <44834743+ad-daniel@users.noreply.github.com>

* Update bin/install_tx2.sh

Co-authored-by: ad-daniel <44834743+ad-daniel@users.noreply.github.com>

* Update install_tx2.sh

* Update install_tx2.sh

* Update install_tx2.sh

* Update install_tx2.sh

* Update install_tx2.sh

* Update edgespeechnets_learner.py

* Update install_tx2.sh

* This PR includes the docker file for tx-2.

Tools that were tested under opendr/projects/perception and work:
- [x] Activity recognition
- [x] Face recognition
- [x] Lightweight Open Pose
- [x] Multimodal human centric
- [x] Object Detection 2D
   - [x] CenterNet
   - [x] Detr
   - [x] Gem
   - [x] RetinaFace
   - [x] SSD
   - [x] YOLOv3
- [x] Object tracking 2d
- [x] Semantic segmentation - Bisenet
- [x] Speech command recognition

The rest of the demos are not complete, i.e. missing an inference demo and/or data to run it.

* This PR includes the docker file for tx-2.

Tools that were tested under opendr/projects/perception and work:
- [x] Activity recognition
- [x] Face recognition
- [x] Lightweight Open Pose
- [x] Multimodal human centric
- [x] Object Detection 2D
   - [x] CenterNet
   - [x] Detr
   - [x] Gem
   - [x] RetinaFace
   - [x] SSD
   - [x] YOLOv3
- [x] Object tracking 2d
- [x] Semantic segmentation - Bisenet
- [x] Speech command recognition

The rest of the demos are not complete, i.e. missing an inference demo and/or data to run it.

* This PR includes the docker file for tx-2.

Tools that were tested under opendr/projects/perception and work:
- [x] Activity recognition
- [x] Face recognition
- [x] Lightweight Open Pose
- [x] Multimodal human centric
- [x] Object Detection 2D
   - [x] CenterNet
   - [x] Detr
   - [x] Gem
   - [x] RetinaFace
   - [x] SSD
   - [x] YOLOv3
- [x] Object tracking 2d
- [x] Semantic segmentation - Bisenet
- [x] Speech command recognition

The rest of the demos are not complete, i.e. missing an inference demo and/or data to run it.

* Added argument for tx2, agx and nx support.
Added instructions on installation.md

* Added argument for tx2, agx and nx support.
Added instructions on installation.md

* Added argument for tx2, agx and nx support.
Added instructions on installation.md

* Update installation.md

* Update activate_nvidia.sh

Corrected python to point to python3 on Nvidia

* Working ROS and correct export of Python version

* Working ROS and correct export of Python version

* Working ROS and correct export of Python version

* Update Dockerfile-embedded

* Update Dockerfile-embedded

* Update Dockerfile-embedded

* Working ROS and correct export of Python version

* Fixed catkin_make, added it to dockerfile.

* Added missing HINT from installation.md

* Updates on default values for FairMOT ros node class ctor

* Fixed duplicate shortcut on deepsort ros node argparse

* Fixed missing shortcut on rgbd hand gesture reco ros node argparse

* Added "opendr_" to data gen package and "_node" to the node file name

* Renamed package to "opendr_perception" and added "_node" to scripts

* Applied fixes to yolov5

* Added "opendr_" to planning package

* Added "opendr_" to bridge package

* Added "opendr_" to simulation package

* Fixed old version of torch in pip_requirements.txt

* Renamed ros bridge package doc

* Renamed dirs, etc. from opendr_ros_bridge to opendr_bridge to match the actual package name

* Fixed ros readme link from customize.md

* Fixed index.md link for opendr-ros-bridge.md

* Minor typo fix in issues.md

* Fixed bridge package name in its doc

* Fixed bridge imports

* Updated for dev_branch

Fixed some missing dependencies

* Fixed some missing dependencies

* add detectron2 (test with new p3.8 Dockerfile)

* Detectron2 issues

* Update install_nvidia.sh

* python3.8 test

* Rolling back to Python3.6.9

* Update install_nvidia.sh

* Updated Dockerfile and dependencies for embedded devices
Changed python to python3 on all ROS nodes
Fixed a couple of issues
Added some exports in activate_nvidia.sh

* Various Fixes

* Removed torch import on data.py and the corresponding reference in Video class.
Added some docker run instructions

* Remove unused import

* Remove Typehint for pyhton compatibilty

* Apply suggestions from code review

Dockerfile-embedded suggested changes

Co-authored-by: Nikolaos Passalis <passalis@users.noreply.github.com>

* Fixed suggested changes

* Format

Co-authored-by: ad-daniel <44834743+ad-daniel@users.noreply.github.com>
Co-authored-by: ad-daniel <daniel.dias@epfl.ch>
Co-authored-by: Nikolaos Passalis <passalis@users.noreply.github.com>
Co-authored-by: tsampazk <tsampaka@csd.auth.gr>
Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com>

Dockerfile-embedded

@@ -0,0 +1,201 @@
+#
+# This dockerfile roughly follows the 'Installing from source' ROS instructions from:
+#   http://wiki.ros.org/noetic/Installation/Source
+#
+ARG BASE_IMAGE=nvcr.io/nvidia/l4t-base:r32.6.1
+FROM ${BASE_IMAGE}
+ARG device=nx
+
+ARG ROS_PKG=ros_base
+ENV ROS_DISTRO=noetic
+ENV ROS_ROOT=/opt/ros/${ROS_DISTRO}
+ENV ROS_PYTHON_VERSION=3
+
+ENV DEBIAN_FRONTEND=noninteractive
+
+ENV CUDA_HOME="/usr/local/cuda"
+ENV PATH="/usr/local/cuda/bin:${PATH}"
+ENV LD_LIBRARY_PATH="/usr/local/cuda/lib64:${LD_LIBRARY_PATH}"
+
+WORKDIR /workspace
+
+#
+# OpenCV - https://github.com/mdegans/nano_build_opencv/blob/master/build_opencv.sh
+#
+ARG OPENCV_VERSION="4.4.0"
+
+# install build dependencies
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+            build-essential \
+            gfortran \
+            cmake \
+            git \
+            file \
+            tar \
+            libatlas-base-dev \
+            libavcodec-dev \
+            libavformat-dev \
+            libavresample-dev \
+            libcanberra-gtk3-module \
+            libdc1394-22-dev \
+            libeigen3-dev \
+            libglew-dev \
+            libgstreamer-plugins-base1.0-dev \
+            libgstreamer-plugins-good1.0-dev \
+            libgstreamer1.0-dev \
+            libgtk-3-dev \
+            libjpeg-dev \
+            libjpeg8-dev \
+            libjpeg-turbo8-dev \
+            liblapack-dev \
+            liblapacke-dev \
+            libopenblas-dev \
+            libpng-dev \
+            libpostproc-dev \
+            libswscale-dev \
+            libtbb-dev \
+            libtbb2 \
+            libtesseract-dev \
+            libtiff-dev \
+            libv4l-dev \
+            libxine2-dev \
+            libxvidcore-dev \
+            libx264-dev \
+            libgtkglext1 \
+            libgtkglext1-dev \
+            pkg-config \
+            qv4l2 \
+            v4l-utils \
+            zlib1g-dev
+
+RUN apt-get install ca-certificates -y
+
+# OpenCV looks for the cuDNN version in cudnn_version.h, but it's been renamed to cudnn_version_v8.h
+RUN ln -s /usr/include/$(uname -i)-linux-gnu/cudnn_version_v8.h /usr/include/$(uname -i)-linux-gnu/cudnn_version.h
+
+# Architecture-specific build options
+ARG CUDA_ARCH_BIN=""
+ARG ENABLE_NEON="OFF"
+
+# Clone and configure OpenCV repo
+RUN git clone --depth 1 --branch ${OPENCV_VERSION} https://github.com/opencv/opencv.git && \
+    git clone --depth 1 --branch ${OPENCV_VERSION} https://github.com/opencv/opencv_contrib.git && \
+    cd opencv && \
+    mkdir build && \
+    cd build && \
+    echo "configuring OpenCV ${OPENCV_VERSION}, CUDA_ARCH_BIN=${CUDA_ARCH_BIN}, ENABLE_NEON=${ENABLE_NEON}" && \
+    cmake \
+    -D CPACK_BINARY_DEB=ON \
+    -D BUILD_EXAMPLES=OFF \
+    -D BUILD_opencv_python2=OFF \
+    -D BUILD_opencv_python3=ON \
+    -D BUILD_opencv_java=OFF \
+    -D CMAKE_BUILD_TYPE=RELEASE \
+    -D CMAKE_INSTALL_PREFIX=/usr/local \
+    -D CUDA_ARCH_BIN=${CUDA_ARCH_BIN} \
+    -D CUDA_ARCH_PTX= \
+    -D CUDA_FAST_MATH=ON \
+    -D CUDNN_INCLUDE_DIR=/usr/include/$(uname -i)-linux-gnu \
+    -D EIGEN_INCLUDE_PATH=/usr/include/eigen3 \
+    -D WITH_EIGEN=ON \
+    -D ENABLE_NEON=${ENABLE_NEON} \
+    -D OPENCV_DNN_CUDA=ON \
+    -D OPENCV_ENABLE_NONFREE=ON \
+    -D OPENCV_EXTRA_MODULES_PATH=/workspace/opencv_contrib/modules \
+    -D OPENCV_GENERATE_PKGCONFIG=ON \
+    -D WITH_CUBLAS=ON \
+    -D WITH_CUDA=ON \
+    -D WITH_CUDNN=ON \
+    -D WITH_GSTREAMER=ON \
+    -D WITH_LIBV4L=ON \
+    -D WITH_OPENGL=ON \
+    -D WITH_OPENCL=OFF \
+    -D WITH_IPP=OFF \
+    -D WITH_TBB=ON \
+    -D BUILD_TIFF=ON \
+    -D BUILD_PERF_TESTS=OFF \
+    -D BUILD_TESTS=OFF \
+    ../
+
+RUN cd opencv/build && make -j$(nproc)
+RUN cd opencv/build && make install
+RUN cd opencv/build && make package
+
+RUN cd opencv/build && tar -czvf OpenCV-${OPENCV_VERSION}-$(uname -i).tar.gz *.deb
+
+RUN apt-get update -y
+RUN apt-get install software-properties-common -y && \
+    apt-get update
+
+# Add the ROS deb repo to the apt sources list
+
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+    git \
+    cmake \
+    build-essential \
+    curl \
+    wget \
+    gnupg2 \
+    lsb-release \
+    ca-certificates \
+    && rm -rf /var/lib/apt/lists/*
+
+RUN sh -c 'echo "deb http://packages.ros.org/ros/ubuntu $(lsb_release -sc) main" > /etc/apt/sources.list.d/ros-latest.list'
+RUN curl -s https://raw.githubusercontent.com/ros/rosdistro/master/ros.asc | apt-key add -
+
+#
+# Install bootstrap dependencies
+#
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+    libpython3-dev \
+    python3-rosdep \
+    python3-rosinstall-generator \
+    python3-vcstool \
+    libyaml-cpp-dev \
+    build-essential && \
+    rosdep init && \
+    rosdep update && \
+    rm -rf /var/lib/apt/lists/*
+
+#
+# Download/build the ROS source
+#
+RUN mkdir ros_catkin_ws && \
+    cd ros_catkin_ws && \
+    # try below for ros-noetic-vision-msgs ros-noetic-geometry-msgs ros-noetic-sensor-msgs ros-noetic-audio-common-msgs \
+    rosinstall_generator ${ROS_PKG} audio_common_msgs --rosdistro ${ROS_DISTRO} --deps --tar > ${ROS_DISTRO}-${ROS_PKG}.rosinstall && \
+    rosinstall_generator ${ROS_PKG} sensor_msgs --rosdistro ${ROS_DISTRO} --deps --tar > ${ROS_DISTRO}-${ROS_PKG}.rosinstall && \
+    rosinstall_generator ${ROS_PKG} geometry_msgs --rosdistro ${ROS_DISTRO} --deps --tar > ${ROS_DISTRO}-${ROS_PKG}.rosinstall && \
+    rosinstall_generator ${ROS_PKG} vision_msgs --rosdistro ${ROS_DISTRO} --deps --tar > ${ROS_DISTRO}-${ROS_PKG}.rosinstall && \
+    mkdir src && \
+    cd src && \
+    git clone https://github.com/ros/resource_retriever && \
+    cd ../ && \
+    vcs import --input ${ROS_DISTRO}-${ROS_PKG}.rosinstall ./src && \
+    apt-get update && \
+    apt-get install -y gir1.2-gstreamer-1.0 && \
+    rosdep install --from-paths ./src --ignore-packages-from-source --rosdistro ${ROS_DISTRO} --skip-keys python3-pykdl -y && \
+    python3 ./src/catkin/bin/catkin_make_isolated --install --install-space ${ROS_ROOT} -DCMAKE_BUILD_TYPE=Release && \
+    rm -rf /var/lib/apt/lists/*
+
+
+RUN echo 'source /opt/ros/${ROS_DISTRO}/setup.bash' >> /root/.bashrc
+CMD ["bash"]
+WORKDIR /
+
+RUN git clone --depth 1 --recurse-submodules -j8 --branch tx2_install_dev https://github.com/opendr-eu/opendr
+RUN apt-get update
+RUN cd ./opendr && ./bin/install_nvidia.sh $device
+RUN cd ./opendr/projects/opendr_ws/src && \
+    git clone --branch noetic https://github.com/ros-perception/vision_opencv && \
+    git clone --branch develop https://github.com/bosch-ros-pkg/usb_cam.git && \
+    cd ./usb_cam && git reset --hard 3ce8ee1 && cd ../ && \
+    git clone https://github.com/ros-perception/image_common.git && \
+    git clone https://github.com/ros-drivers/audio_common && \
+    sed -i 's/(Boost REQUIRED python37)/(Boost REQUIRED python3)/' ./vision_opencv/cv_bridge/CMakeLists.txt && \
+    cd ../
+
+RUN /bin/bash -c '. /opt/ros/noetic/setup.bash; cd /opendr/projects/opendr_ws; catkin_make'

bin/activate_nvidia.sh

@@ -0,0 +1,12 @@
+#!/bin/sh
+export OPENDR_HOME=$PWD
+export PYTHONPATH=$OPENDR_HOME/src:$PYTHONPATH
+alias python=python3
+export LD_LIBRARY_PATH=$OPENDR_HOME/lib:$LD_LIBRARY_PATH
+
+export PATH=/usr/local/cuda/bin:$PATH
+export MXNET_HOME=$OPENDR_HOME/mxnet/
+export PYTHONPATH=$MXNET_HOME/python:$PYTHONPATH
+export MXNET_CUDNN_AUTOTUNE_DEFAULT=0
+export LC_ALL="C.UTF-8"
+export MPLBACKEND=TkAgg