slam-indoor-code

Repository with source code of SLAM indoor project

Requirements

OpenCV 4.8.0
CMake 3.26.2
Ceres-solver 2.2.0
Optional: CUDA 12.3.2

Environment configuration

OpenCV installation

Install dependencies and tools:

sudo apt update
sudo apt install cmake libtbb2 g++ wget unzip ffmpeg libgtk2.0-dev libavformat-dev libavcodec-dev libavutil-dev libswscale-dev libtbb-dev libjpeg-dev libpng-dev libtiff-dev
sudo apt install libvtk7-dev
sudo apt install build-essential
sudo apt install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev
#тут некоторые библиотеки могли не установится но пофиг
sudo apt install libeigen3-dev libgflags-dev libgoogle-glog-dev libatlas-base-dev libsuitesparse-
#тут вроде всё должно сработать

Install Ceres-Solver
Optional: Install CUDA
Download sources

wget -O opencv.zip https://github.com/opencv/opencv/archive/4.8.0.zip
unzip opencv.zip  # files will be extracted to ./opencv-4.8.0

wget -O opencv-contrib.zip https://github.com/opencv/opencv_contrib/archive/refs/tags/4.8.0.zip
unzip opencv-contrib.zip  # files will be extracted to ./opencv_contrib-4.8.0

mkdir opencv-build
cd opencv-build

Build & install
- See CUDA build options here

cmake -D CMAKE_BUILD_TYPE=RELEASE \
-D CMAKE_INSTALL_PREFIX=/usr/local \
-D OPENCV_EXTRA_MODULES_PATH=../opencv_contrib-4.8.0/modules/ \
-D BUILD_SHARED_LIBS=ON \
-D BUILD_opencv_sfm=ON \
-D OPENCV_ENABLE_NONFREE=ON \
-D BUILD_SHARED_LIBS=ON \
-D BUILD_TESTS=ON \
-D OPENCV_GENERATE_PKGCONFIG=ON \
-D BUILD_EXAMPLES=ON \
-D WITH_QT=ON \
-D WITH_GTK=ON \
-D WITH_OPENGL=ON \
-D WITH_FFMPEG=ON \
-D WITH_TBB=ON \
-D WITH_V4L=ON \
-D WITH_VTK=ON \
../opencv-4.8.0/  

# Make sure FFMPEG and its modules marked "YES"

make -j8  # Number of jobs can be specified
sudo make install

Ceres installation

sudo apt-get install libeigen3-dev libgflags-dev libgoogle-glog-dev

wget http://ceres-solver.org/ceres-solver-2.2.0.tar.gz
tar zxf ceres-solver-2.2.0.tar.gz
mkdir ceres-bin
cd ceres-bin
cmake ../ceres-solver-2.2.0

make -j3
make test
sudo make install

Run to test:

./bin/simple_bundle_adjuster ../ceres-solver-2.2.0/data/problem-16-22106-pre.txt

CUDA installation

CUDA packages

wget https://developer.download.nvidia.com/compute/cuda/repos/wsl-ubuntu/x86_64/cuda-keyring_1.1-1_all.deb
sudo dpkg -i cuda-keyring_1.1-1_all.deb
sudo apt-get update
sudo apt-get -y install cuda-toolkit-12-3

CUDA build

Specify version for -D CUDA_ARCH_BIN=<version> ragrding to your hardware using this site

cmake -D CMAKE_BUILD_TYPE=RELEASE \
-D CMAKE_INSTALL_PREFIX=/usr/local \
-D OPENCV_EXTRA_MODULES_PATH=../opencv_contrib-4.8.0/modules/ \
-D BUILD_SHARED_LIBS=ON \
-D BUILD_opencv_sfm=ON \
-D OPENCV_ENABLE_NONFREE=ON \
-D BUILD_SHARED_LIBS=ON \
-D BUILD_TESTS=ON \
-D OPENCV_GENERATE_PKGCONFIG=ON \
-D BUILD_EXAMPLES=ON \
-D WITH_QT=ON \
-D WITH_GTK=ON \
-D WITH_OPENGL=ON \
-D WITH_FFMPEG=ON \
-D WITH_TBB=ON \
-D WITH_V4L=ON \
-D WITH_VTK=ON \
-D ENABLE_FAST_MATH=1 \
-D CUDA_FAST_MATH=1 \
-D WITH_CUBLAS=1 \
-D WITH_CUDA=ON \
-D BUILD_opencv_cudacodec=OFF \
-D WITH_CUDNN=ON \
-D OPENCV_DNN_CUDA=OFF \
-D CUDA_ARCH_BIN=<version> \
-D WITH_GSTREAMER=ON \
../opencv-4.8.0/  

# Make sure FFMPEG, its and CUDA (not cuDNN) modules marked "YES"

Configuration files (may be outdated)

Install nlohmann-json:

sudo apt install nlohmann-json3-dev

JSON config example (path to it should be specified as command line argument):

{
  "onlyViz": true,

  "calibrate": false,
  "visualCalibration": true,
  "calibrationPath": "./config/samsung-hv-2.xml",

  "usePhotosCycle": false,
  //  "photosPathPattern": "/mnt/c/Users/bakug/YandexDisk/NSU/private/2-1/PAK/static/photos/samsung-room-new/_ (*).JPG",
  "photosPathPattern": "/mnt/c/Users/bakug/YandexDisk/NSU/private/2-1/PAK/static/photos/samsung-4to3-tumbochka/*.JPG",
  // "videoSourcePath" has an effect when "usePhotosCycle" is false
  "videoSourcePath": "/mnt/c/Users/bakug/YandexDisk/NSU/private/2-1/PAK/static/samsung-new-tumbochka-fhd.MP4",

  // If "onlyViz" is true program searches files with data in this folder
  "outputDataDir": "./data/video_report/table-photos-CPU",

  "threadsCount": 1,

  "useUndistortion": false,

  "requiredExtractedPointsCount": 10000,
  "featureExtractingThreshold": 1,

  "framesBatchSize": 210,
  "skipFramesFromBatchHead": 0,
  "useFirstFitInBatch": true,

  "requiredMatchedPointsCount": 500,

  // This block has an effect when "useFeatureTracker" is false
  "useFM-SIFT-FLANN": true, // NORM_L2 in CUDA
  "useFM-SIFT-BF": false, // NORM_L1 in CUDA
  "useFM-ORB": false, // NORM_HAMMING in CUDA

  "knnMatcherDistance": 0.7,

  // Now this parameters are used only for the first pair of frames
  "RPUseRANSAC": true,
  "RPRANSACProb": 0.999,
  "RPRANSACThreshold": 5.0,
  "RPDistanceThreshold": 200.0,


  "useBundleAdjustment": false,
  // Next parameter also specifes max frames datas cnt which will be uploaded to global data at the one moment
  "BAMaxFramesCnt": 8,
  "BAThreadsCnt": 12,

  // Priority of loss functions' flags is from top to bottom
  "BAUseTrivialLossFunction": false,

  "BAUseHuberLossFunction": true,
  "BAHuberLossFunctionParameter": 4.0,

  "BAUseCauchyLossFunction": false,
  "BACauchyLossFunctionParameter": 4.0,

  "BAUseArctanLossFunction": false,
  "BAArctanLossFunctionParameter": 2.0,

  "BAUseTukeyLossFunction": false,
  "BATukeyLossFunctionParameter": 4.0
}

So now you can specify program working using configs and run using ./rebuild_and_run.sh <path/to/config.json (write chmod a+x rebuild_and_run.sh to make this file executable)

For launching with CUDA use rebuild_and_run_cuda.sh

Build our binary

Simple building:

cmake . -B build
make -C build -j8

CUDA building:

cmake . -B build -D USE_CUDA=YES
make -C build -j8

So now you can execute ./build/slam-indoor-code <path/to/config.json>

FIT-2023-SLAM-indoor/slam-indoor-code