[
{
"path": ".gitignore",
"content": "cis565_getting_started\ncis565_getting_started_generated_kernel*\n*.orig\n*.filters\n*.sln\n*.vcxproj\n*.xcodeproj\nbuild\n\n# Created by https://www.gitignore.io/api/linux,osx,sublimetext,windows,jetbrains,vim,emacs,cmake,c++,cuda,visualstudio,webstorm,eclipse,xcode\n\n### Linux ###\n*~\n\n# KDE directory preferences\n.directory\n\n# Linux trash folder which might appear on any partition or disk\n.Trash-*\n\n\n### OSX ###\n.DS_Store\n.AppleDouble\n.LSOverride\n\n# Icon must end with two \\r\nIcon\n\n\n# Thumbnails\n._*\n\n# Files that might appear in the root of a volume\n.DocumentRevisions-V100\n.fseventsd\n.Spotlight-V100\n.TemporaryItems\n.Trashes\n.VolumeIcon.icns\n\n# Directories potentially created on remote AFP share\n.AppleDB\n.AppleDesktop\nNetwork Trash Folder\nTemporary Items\n.apdisk\n\n\n### SublimeText ###\n# cache files for sublime text\n*.tmlanguage.cache\n*.tmPreferences.cache\n*.stTheme.cache\n\n# workspace files are user-specific\n*.sublime-workspace\n\n# project files should be checked into the repository, unless a significant\n# proportion of contributors will probably not be using SublimeText\n# *.sublime-project\n\n# sftp configuration file\nsftp-config.json\n\n\n### Windows ###\n# Windows image file caches\nThumbs.db\nehthumbs.db\n\n# Folder config file\nDesktop.ini\n\n# Recycle Bin used on file shares\n$RECYCLE.BIN/\n\n# Windows Installer files\n*.cab\n*.msi\n*.msm\n*.msp\n\n# Windows shortcuts\n*.lnk\n\n\n### JetBrains ###\n# Covers JetBrains IDEs: IntelliJ, RubyMine, PhpStorm, AppCode, PyCharm, CLion, Android Studio\n\n*.iml\n\n## Directory-based project format:\n#.idea/\n# if you remove the above rule, at least ignore the following:\n\n# User-specific stuff:\n.idea/workspace.xml\n.idea/tasks.xml\n.idea/dictionaries\n\n# Sensitive or high-churn files:\n.idea/dataSources.ids\n.idea/dataSources.xml\n.idea/sqlDataSources.xml\n.idea/dynamic.xml\n.idea/uiDesigner.xml\n\n# Gradle:\n.idea/gradle.xml\n.idea/libraries\n\n# Mongo Explorer plugin:\n.idea/mongoSettings.xml\n\n## File-based project format:\n*.ipr\n*.iws\n\n## Plugin-specific files:\n\n# IntelliJ\n/out/\n\n# mpeltonen/sbt-idea plugin\n.idea_modules/\n\n# JIRA plugin\natlassian-ide-plugin.xml\n\n# Crashlytics plugin (for Android Studio and IntelliJ)\ncom_crashlytics_export_strings.xml\ncrashlytics.properties\ncrashlytics-build.properties\n\n\n### Vim ###\n[._]*.s[a-w][a-z]\n[._]s[a-w][a-z]\n*.un~\nSession.vim\n.netrwhist\n*~\n\n\n### Emacs ###\n# -*- mode: gitignore; -*-\n*~\n\\#*\\#\n/.emacs.desktop\n/.emacs.desktop.lock\n*.elc\nauto-save-list\ntramp\n.\\#*\n\n# Org-mode\n.org-id-locations\n*_archive\n\n# flymake-mode\n*_flymake.*\n\n# eshell files\n/eshell/history\n/eshell/lastdir\n\n# elpa packages\n/elpa/\n\n# reftex files\n*.rel\n\n# AUCTeX auto folder\n/auto/\n\n# cask packages\n.cask/\n\n\n### CMake ###\nCMakeCache.txt\nCMakeFiles\nCMakeScripts\nMakefile\ncmake_install.cmake\ninstall_manifest.txt\n\n\n### C++ ###\n# Compiled Object files\n*.slo\n*.lo\n*.o\n*.obj\n\n# Precompiled Headers\n*.gch\n*.pch\n\n# Compiled Dynamic libraries\n*.so\n*.dylib\n*.dll\n\n# Fortran module files\n*.mod\n\n# Compiled Static libraries\n*.lai\n*.la\n*.a\n*.lib\n!/external/lib/*/**/*\n\n# Executables\n*.exe\n*.out\n*.app\n\n\n### CUDA ###\n*.i\n*.ii\n*.gpu\n*.ptx\n*.cubin\n*.fatbin\n\n\n### VisualStudio ###\n## Ignore Visual Studio temporary files, build results, and\n## files generated by popular Visual Studio add-ons.\n\n# User-specific files\n*.suo\n*.user\n*.userosscache\n*.sln.docstates\n\n# User-specific files (MonoDevelop/Xamarin Studio)\n*.userprefs\n\n# Build results\n[Dd]ebug/\n[Dd]ebugPublic/\n[Rr]elease/\n[Rr]eleases/\nx64/\nx86/\nbuild/\nbld/\n[Bb]in/\n[Oo]bj/\n\n# Visual Studio 2015 cache/options directory\n.vs/\n# Uncomment if you have tasks that create the project's static files in wwwroot\n#wwwroot/\n\n# MSTest test Results\n[Tt]est[Rr]esult*/\n[Bb]uild[Ll]og.*\n\n# NUNIT\n*.VisualState.xml\nTestResult.xml\n\n# Build Results of an ATL Project\n[Dd]ebugPS/\n[Rr]eleasePS/\ndlldata.c\n\n# DNX\nproject.lock.json\nartifacts/\n\n*_i.c\n*_p.c\n*_i.h\n*.ilk\n*.meta\n*.obj\n*.pch\n*.pdb\n*.pgc\n*.pgd\n*.rsp\n*.sbr\n*.tlb\n*.tli\n*.tlh\n*.tmp\n*.tmp_proj\n*.log\n*.vspscc\n*.vssscc\n.builds\n*.pidb\n*.svclog\n*.scc\n\n# Chutzpah Test files\n_Chutzpah*\n\n# Visual C++ cache files\nipch/\n*.aps\n*.ncb\n*.opensdf\n*.sdf\n*.cachefile\n\n# Visual Studio profiler\n*.psess\n*.vsp\n*.vspx\n\n# TFS 2012 Local Workspace\n$tf/\n\n# Guidance Automation Toolkit\n*.gpState\n\n# ReSharper is a .NET coding add-in\n_ReSharper*/\n*.[Rr]e[Ss]harper\n*.DotSettings.user\n\n# JustCode is a .NET coding add-in\n.JustCode\n\n# TeamCity is a build add-in\n_TeamCity*\n\n# DotCover is a Code Coverage Tool\n*.dotCover\n\n# NCrunch\n_NCrunch_*\n.*crunch*.local.xml\nnCrunchTemp_*\n\n# MightyMoose\n*.mm.*\nAutoTest.Net/\n\n# Web workbench (sass)\n.sass-cache/\n\n# Installshield output folder\n[Ee]xpress/\n\n# DocProject is a documentation generator add-in\nDocProject/buildhelp/\nDocProject/Help/*.HxT\nDocProject/Help/*.HxC\nDocProject/Help/*.hhc\nDocProject/Help/*.hhk\nDocProject/Help/*.hhp\nDocProject/Help/Html2\nDocProject/Help/html\n\n# Click-Once directory\npublish/\n\n# Publish Web Output\n*.[Pp]ublish.xml\n*.azurePubxml\n# TODO: Comment the next line if you want to checkin your web deploy settings \n# but database connection strings (with potential passwords) will be unencrypted\n*.pubxml\n*.publishproj\n\n# NuGet Packages\n*.nupkg\n# The packages folder can be ignored because of Package Restore\n**/packages/*\n# except build/, which is used as an MSBuild target.\n!**/packages/build/\n# Uncomment if necessary however generally it will be regenerated when needed\n#!**/packages/repositories.config\n\n# Windows Azure Build Output\ncsx/\n*.build.csdef\n\n# Windows Store app package directory\nAppPackages/\n\n# Visual Studio cache files\n# files ending in .cache can be ignored\n*.[Cc]ache\n# but keep track of directories ending in .cache\n!*.[Cc]ache/\n\n# Others\nClientBin/\n[Ss]tyle[Cc]op.*\n~$*\n*~\n*.dbmdl\n*.dbproj.schemaview\n*.pfx\n*.publishsettings\nnode_modules/\norleans.codegen.cs\n\n# RIA/Silverlight projects\nGenerated_Code/\n\n# Backup & report files from converting an old project file\n# to a newer Visual Studio version. Backup files are not needed,\n# because we have git ;-)\n_UpgradeReport_Files/\nBackup*/\nUpgradeLog*.XML\nUpgradeLog*.htm\n\n# SQL Server files\n*.mdf\n*.ldf\n\n# Business Intelligence projects\n*.rdl.data\n*.bim.layout\n*.bim_*.settings\n\n# Microsoft Fakes\nFakesAssemblies/\n\n# Node.js Tools for Visual Studio\n.ntvs_analysis.dat\n\n# Visual Studio 6 build log\n*.plg\n\n# Visual Studio 6 workspace options file\n*.opt\n\n# Visual Studio LightSwitch build output\n**/*.HTMLClient/GeneratedArtifacts\n**/*.DesktopClient/GeneratedArtifacts\n**/*.DesktopClient/ModelManifest.xml\n**/*.Server/GeneratedArtifacts\n**/*.Server/ModelManifest.xml\n_Pvt_Extensions\n\n\n### WebStorm ###\n# Covers JetBrains IDEs: IntelliJ, RubyMine, PhpStorm, AppCode, PyCharm, CLion, Android Studio\n\n*.iml\n\n## Directory-based project format:\n.idea/\n# if you remove the above rule, at least ignore the following:\n\n# User-specific stuff:\n# .idea/workspace.xml\n# .idea/tasks.xml\n# .idea/dictionaries\n\n# Sensitive or high-churn files:\n# .idea/dataSources.ids\n# .idea/dataSources.xml\n# .idea/sqlDataSources.xml\n# .idea/dynamic.xml\n# .idea/uiDesigner.xml\n\n# Gradle:\n# .idea/gradle.xml\n# .idea/libraries\n\n# Mongo Explorer plugin:\n# .idea/mongoSettings.xml\n\n## File-based project format:\n*.ipr\n*.iws\n\n## Plugin-specific files:\n\n# IntelliJ\n/out/\n\n# mpeltonen/sbt-idea plugin\n.idea_modules/\n\n# JIRA plugin\natlassian-ide-plugin.xml\n\n# Crashlytics plugin (for Android Studio and IntelliJ)\ncom_crashlytics_export_strings.xml\ncrashlytics.properties\ncrashlytics-build.properties\n\n\n### Eclipse ###\n*.pydevproject\n.metadata\n.gradle\nbin/\ntmp/\n*.tmp\n*.bak\n*.swp\n*~.nib\nlocal.properties\n.settings/\n.loadpath\n\n# Eclipse Core\n.project\n\n# External tool builders\n.externalToolBuilders/\n\n# Locally stored \"Eclipse launch configurations\"\n*.launch\n\n# CDT-specific\n.cproject\n\n# JDT-specific (Eclipse Java Development Tools)\n.classpath\n\n# Java annotation processor (APT)\n.factorypath\n\n# PDT-specific\n.buildpath\n\n# sbteclipse plugin\n.target\n\n# TeXlipse plugin\n.texlipse\n\n\n### Xcode ###\n# Xcode\n#\n# gitignore contributors: remember to update Global/Xcode.gitignore, Objective-C.gitignore & Swift.gitignore\n\n## Build generated\nbuild/\nDerivedData\n\n## Various settings\n*.pbxuser\n!default.pbxuser\n*.mode1v3\n!default.mode1v3\n*.mode2v3\n!default.mode2v3\n*.perspectivev3\n!default.perspectivev3\nxcuserdata\n\n## Other\n*.xccheckout\n*.moved-aside\n*.xcuserstate\n"
},
{
"path": "CMakeLists.txt",
"content": "cmake_minimum_required(VERSION 3.1)\nproject(cis565_ScanMatching)\n\nset_property(GLOBAL PROPERTY USE_FOLDERS ON)\n\nset(CMAKE_MODULE_PATH \"${CMAKE_SOURCE_DIR}/cmake\" ${CMAKE_MODULE_PATH})\n\nset(CMAKE_ARCHIVE_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/lib)\nset(CMAKE_LIBRARY_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/lib)\nset(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/bin)\n\n# Enable C++11 for host code\nset(CMAKE_CXX_STANDARD 11)\n\n# Set a default build type if none was specified\nif(NOT CMAKE_BUILD_TYPE AND NOT CMAKE_CONFIGURATION_TYPES)\n SET(CMAKE_BUILD_TYPE Release CACHE STRING \"Choose the type of build.\" FORCE)\n # Set the possible values of build type for cmake-gui\n SET_PROPERTY(CACHE CMAKE_BUILD_TYPE PROPERTY STRINGS \"Debug\" \"Release\" \"MinSizeRel\" \"RelWithDebInfo\")\nendif()\n\n########################################\n# CUDA Setup\n########################################\nfind_package(CUDA 10 REQUIRED)\ninclude(${CMAKE_MODULE_PATH}/CUDAComputesList.cmake)\n\nlist(APPEND CUDA_NVCC_FLAGS ${CUDA_GENERATE_CODE})\nlist(APPEND CUDA_NVCC_FLAGS_DEBUG \"-g -G\")\nset(CUDA_VERBOSE_BUILD ON)\n\nif(WIN32)\n # Set up include and lib paths\n set(CUDA_HOST_COMPILER ${CMAKE_CXX_COMPILER} CACHE FILEPATH \"Host side compiler used by NVCC\" FORCE)\nendif(WIN32)\n########################################\n\nfind_package(OpenGL REQUIRED)\n\nif(UNIX)\n find_package(glfw3 REQUIRED)\n find_package(GLEW REQUIRED)\n set(LIBRARIES glfw ${GLEW_LIBRARIES} ${OPENGL_gl_LIBRARY})\nelse(UNIX)\n set(EXTERNAL \"external\")\n\n set(GLFW_ROOT_DIR ${EXTERNAL})\n set(GLFW_USE_STATIC_LIBS ON)\n find_package(GLFW REQUIRED)\n\n set(GLEW_ROOT_DIR ${EXTERNAL})\n set(GLEW_USE_STATIC_LIBS ON)\n find_package(GLEW REQUIRED)\n\n add_definitions(${GLEW_DEFINITIONS})\n include_directories(${GLEW_INCLUDE_DIR} ${GLFW_INCLUDE_DIR})\n set(LIBRARIES ${GLEW_LIBRARY} ${GLFW_LIBRARY} ${OPENGL_LIBRARY})\nendif(UNIX)\n\nset(GLM_ROOT_DIR \"external\")\nfind_package(GLM REQUIRED)\ninclude_directories(${GLM_INCLUDE_DIRS})\n\nset(headers\n src/cudaMat4.hpp\n src/glslUtility.hpp\n src/scanmatch.h\n src/octree.h\n src/main.hpp\n src/pointcloud.h\n src/utilityCore.hpp\n )\n\nset(sources\n src/glslUtility.cpp\n src/scanmatch.cu\n src/octree.cu\n src/main.cpp\n src/pointcloud.cu\n src/utilityCore.cpp\n )\n\nlist(SORT headers)\nlist(SORT sources)\n\nsource_group(Headers FILES ${headers})\nsource_group(Sources FILES ${sources})\n\ncuda_add_executable(${CMAKE_PROJECT_NAME} ${sources} ${headers})\ntarget_link_libraries(${CMAKE_PROJECT_NAME} ${LIBRARIES})\n\nadd_custom_command(\n TARGET ${CMAKE_PROJECT_NAME}\n PRE_BUILD\n COMMAND ${CMAKE_COMMAND} -E copy_directory\n ${CMAKE_SOURCE_DIR}/shaders\n ${CMAKE_BINARY_DIR}/shaders\n )\n"
},
{
"path": "GNUmakefile",
"content": "CMAKE_ALT1 := /usr/local/bin/cmake\nCMAKE_ALT2 := /Applications/CMake.app/Contents/bin/cmake\nCMAKE := $(shell \\\n\twhich cmake 2>/dev/null || \\\n\t([ -e ${CMAKE_ALT1} ] && echo \"${CMAKE_ALT1}\") || \\\n\t([ -e ${CMAKE_ALT2} ] && echo \"${CMAKE_ALT2}\") \\\n\t)\n\nall: Release\n\n\nDebug: build\n\t(cd build && ${CMAKE} -DCMAKE_BUILD_TYPE=$@ .. && make)\n\nMinSizeRel: build\n\t(cd build && ${CMAKE} -DCMAKE_BUILD_TYPE=$@ .. && make)\n\nRelease: build\n\t(cd build && ${CMAKE} -DCMAKE_BUILD_TYPE=$@ .. && make)\n\nRelWithDebugInfo: build\n\t(cd build && ${CMAKE} -DCMAKE_BUILD_TYPE=$@ .. && make)\n\n\nbuild:\n\tmkdir -p build\n\nclean:\n\t((cd build && make clean) 2>&- || true)\n\n.PHONY: all Debug MinSizeRel Release RelWithDebugInfo clean\n"
},
{
"path": "LICENSE",
"content": "MIT License\n\nCopyright (c) 2020 Dhruv Karthik\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof this software and associated documentation files (the \"Software\"), to deal\nin the Software without restriction, including without limitation the rights\nto use, copy, modify, merge, publish, distribute, sublicense, and/or sell\ncopies of the Software, and to permit persons to whom the Software is\nfurnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all\ncopies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\nIMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\nAUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\nLIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\nSOFTWARE.\n"
},
{
"path": "README.md",
"content": "CUDA Scan Matcher via ICP - Octree Optimized \n===============================================================================\n\n**University of Pennsylvania, CIS 565: GPU Programming and Architecture**\n\nDhruv Karthik: [LinkedIn](https://www.linkedin.com/in/dhruvkarthik/)\n\nTested on: Windows 10 Home, Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz, 16GM, GTX 2070 - Compute Capability 7.5\n____________________________________________________________________________________\n   \n____________________________________________________________________________________\n\n ![](\"img/waymotrue.gif\")
\n
\n\nTable of contents\n=================\n * [Scan Matching Algorithm](#scan-matching-algorithm)\n * [Results](#results)\n * [Octree Optimization](#sparse-octree-optimization)\n * [Performance Analysis](#performance-analysis)\n * [Build Instructions](#build-instructions)\n * [Running The Code](#running-the-code)\n \n# Scan Matching Algorithm\nAn intuitive understanding of Scan Matching is best illustrated in this 11 second [video](https://www.youtube.com/watch?v=uzOCS_gdZuM).\nScan Matching seeks to align two similar pointclouds by finding the transformation between them. It does so via the iterative closest point algorithm outlined below in pseudocode:\n```bash\n#Aligns pointcloud A to pointcloud B\ndef scan_match(pointcloud A, pointcloud B):\n 1. For each point in A, find the closest point in B\n 2. Compute a 3-D transformation matrix that aligns the points (Use SVD & Least Squares Regression)\n 3. Update all points in the target by the transformation matrix\n 4. Repeat steps 1-3 until some epsilon convergence\n RETURN : Some Transformation matrix T\n```\nThe key benefit from CUDA is in step 1. Finding the nearest neighbors synchronously takes far longer than finding nearest neighbors via some optimized datastructure like an octree. I implemented 3 versions: A CPU version, a Naive CUDA version, and an Octree version. \n\n# Results\n\n ![](\"img/bunnytrue.gif\")
\n
\n\n\n ![](\"img/buddhatrue.gif\")
\n
\n\n\n ![](\"img/dragontrue.gif\")
\n
\n\n# Sparse Octree Optimization\nA sparse octree is an optimized datastructure that allows for fast nearest neighbor calculations. It is represented as a tree in which each internal node has exactly eight children. We use octrees in order to subdivide our 3D space, ensuring that no *octant* has more than *n* points. If some node has more than *n* vertices, we recursively subdivide this node by splitting it into eight children, and redistributing the points among these 8 children. This speeds up nearest neighbor searches as now, we only need to search for neighbors within our octant. \n \n ![](\"https://developer.nvidia.com/sites/all/modules/custom/gpugems/books/GPUGems2/elementLinks/37_octree_03.jpg\")
\n
\n \n # Performance Analysis\n In order to test the varying number of points, I generated a 3D sinusoid with various widths, thereby increasing the number of points. \n ## Increasing speed of Nearest Neighbor Search\n The table below illustrates the effect of each algorithm on the speed of nearest neighbor search. Clearly, the GPU-Octree performs NN the fastest, as it iterates only through points inside its octant. The Naive GPU iterates through all the points, but does so paralelly, and is therefore faster than the CPU that does so in a triple for loop sequentially.\n\n**Number of points vs Time Taken (ms) for Nearest Neighbor Search (lower is better)**\n \n| Num Points \t| CPU \t| GPU-Naive \t| GPU-Octree \t|\n|------------\t|--------\t|-----------\t|------------\t|\n| 5 \t| 0 \t| 5 \t| 8 \t|\n| 500 \t| 1268 \t| 8 \t| 3 \t|\n| 2500 \t| 23000 \t| 573 \t| 164 \t|\n| 5000 \t| 94000 \t| 903 \t| 193 \t|\n| 250000 \t| 861000 \t| 14000 \t| 400 \t|\n\n ## Effect of Increasing NN on framerate\n Below, we clearly see that the decrease in time across the algorithm directly contributes to an increase in framerate. Astonishingly, we see an over 1000% improvement over the CPU algorithm:\n \n ![](\"img/chart1.png\")
\n