More updates

CMakeLists.txt

@@ -1,4 +1,4 @@
-cmake_minimum_required(VERSION 2.6.0)
+cmake_minimum_required(VERSION 3.22)
 
 project(ElasticFusion)
 

Core/CMakeLists.txt

@@ -1,4 +1,4 @@
-cmake_minimum_required(VERSION 2.6.0)
+cmake_minimum_required(VERSION 3.22)
 
 project(libefusion)
 
@@ -25,7 +25,7 @@ endforeach(NVCC_ARCH)
 include("${CMAKE_MODULE_PATH}/CudaComputeTargetFlags.cmake")
 APPEND_TARGET_ARCH_FLAGS()
 
-set(CUDA_NVCC_FLAGS ${CUDA_NVCC_FLAGS}  "-Xcompiler;-fPIC;--expt-relaxed-constexpr")
+set(CUDA_NVCC_FLAGS ${CUDA_NVCC_FLAGS}  "-Xcompiler;-fPIC;--expt-relaxed-constexpr;--disable-warnings")
 
 CUDA_COMPILE(cuda_objs ${cuda})
 

Core/Deformation.cpp

@@ -30,7 +30,9 @@ Deformation::Deformation()
       graphPosePoints(new std::vector<Eigen::Vector3d>),
       lastDeformTime(0) {
   // x, y, z and init time
-  memset(&rawSampledNodes_w[0], 0, bufferSize);
+  for (int i = 0; i < bufferSize; i++) {
+    rawSampledNodes_w[i] = Eigen::Vector4f::Zero();
+  }
 
   glGenTransformFeedbacks(1, &fid);
   glGenBuffers(1, &vbo);

Core/Ferns.h

@@ -100,12 +100,16 @@ class Ferns {
 
       if (verts) {
         this->initVerts = new Eigen::Vector4f[numPixels];
-        memcpy(this->initVerts, verts, numPixels * sizeof(Eigen::Vector4f));
+        for (int i = 0; i < numPixels; i++) {
+          this->initVerts[i] = verts[i];
+        }
       }
 
       if (norms) {
         this->initNorms = new Eigen::Vector4f[numPixels];
-        memcpy(this->initNorms, norms, numPixels * sizeof(Eigen::Vector4f));
+        for (int i = 0; i < numPixels; i++) {
+          this->initNorms[i] = norms[i];
+        }
       }
     }
 

Core/GlobalModel.cpp

@@ -679,7 +679,9 @@ Eigen::Vector4f* GlobalModel::downloadMap() {
 
   Eigen::Vector4f* vertices = new Eigen::Vector4f[count * 3];
 
-  memset(&vertices[0], 0, count * Vertex::SIZE);
+  for(uint32_t i = 0; i < count * 3; i++) {
+    vertices[i] = Eigen::Vector4f::Zero();
+  }
 
   GLuint downloadVbo;
 

MainController.cpp

@@ -69,7 +69,7 @@ MainController::MainController(int argc, char* argv[])
   confidence = 10.0f;
   depth = 3.0f;
   icp = 10.0f;
-  icpErrThresh = 5e-05;
+  icpErrThresh = 4e-05;
   covThresh = 1e-05;
   photoThresh = 115;
   fernThresh = 0.3095f;

README.md

@@ -12,76 +12,36 @@ Please cite this work if you make use of our system in any of your own endeavors
 
 ## 1.1. Ubuntu ##
 
-* Ubuntu 14.04, 15.04 or 16.04 (Though many other linux distros will work fine)
-* CMake
-* OpenGL
-* [CUDA >= 7.0](https://developer.nvidia.com/cuda-downloads)
-* [OpenNI2](https://github.com/occipital/OpenNI2)
-* SuiteSparse
-* Eigen
-* zlib
-* libjpeg
-* [Pangolin](https://github.com/stevenlovegrove/Pangolin)
-* [librealsense] (https://github.com/IntelRealSense/librealsense) - Optional (for Intel RealSense cameras)
-
-Firstly, add [nVidia's official CUDA repository](https://developer.nvidia.com/cuda-downloads) to your apt sources, then run the following command to pull in most dependencies from the official repos:
+Ubuntu 22.04 on Xorg, NVIDIA drivers 510.73.05, CUDA driver 11.6, CUDA toolkit 11.5 (essentially whatever is in the Ubuntu repos).
 
 ```bash
-sudo apt-get install -y cmake-qt-gui git build-essential libusb-1.0-0-dev libudev-dev openjdk-7-jdk freeglut3-dev libglew-dev cuda-7-5 libsuitesparse-dev libeigen3-dev zlib1g-dev libjpeg-dev
+sudo apt install -y cmake-qt-gui git build-essential libusb-1.0-0-dev libudev-dev openjdk-11-jdk freeglut3-dev libglew-dev libsuitesparse-dev zlib1g-dev libjpeg-dev
 ```
 
-Afterwards install [OpenNI2](https://github.com/occipital/OpenNI2) and [Pangolin](https://github.com/stevenlovegrove/Pangolin) from source. Note, you may need to manually tell CMake where OpenNI2 is since Occipital's fork does not have an install option. It is important to build Pangolin last so that it can find some of the libraries it has optional dependencies on. 
-
-When you have all of the dependencies installed, build the Core followed by the GUI. 
-
-## 1.2. Windows - Visual Studio ##
-* Windows 7/10 with Visual Studio 2013 Update 5 (Though other configurations may work)
-* [CMake] (https://cmake.org/)
-* OpenGL
-* [CUDA >= 7.0](https://developer.nvidia.com/cuda-downloads)
-* [OpenNI2](https://github.com/occipital/OpenNI2)
-* [SuiteSparse] (https://github.com/jlblancoc/suitesparse-metis-for-windows)
-* [Eigen] (http://eigen.tuxfamily.org)
-* [Pangolin](https://github.com/stevenlovegrove/Pangolin)
-  * zlib (Pangolin can automatically download and build this)
-  * libjpeg (Pangolin can automatically download and build this)
-* [librealsense] (https://github.com/IntelRealSense/librealsense) - Optional (for Intel RealSense cameras)
-
-Firstly install cmake and cuda. Then download and build from source OpenNI2, SuiteSparse. Next download Eigen (no need to build it since it is a header-only library). Then download and build from source Pangolin but pay attention to the following cmake settings. There will be a lot of dependencies where path was not found. That is OK except OPENNI2 and EIGEN3 (those should be set to valid paths). You also need to set MSVC_USE_STATIC_CRT to false in order to correctly link to ElasticFusion projects. Also, you can set BUILD_EXAMPLES to false since we don't need them and some were crashing on my machine.
-
-Finally, build Core and GUI.
-
-
-# 2. Is there an easier way to build it? #
-Yes, if you run the *build.sh* script on a fresh clean install of Ubuntu 14.04, 15.04, or 16.04, enter your password for sudo a few times and wait a few minutes all dependencies will get downloaded and installed and it should build everything correctly. This has not been tested on anything but fresh installs, so I would advise using it with caution if you already have some of the dependencies installed.
-
-# 3. Installation issues #
-
-***`#include <Eigen/Core>` not found***
-
 ```bash
-sudo ln -sf /usr/include/eigen3/Eigen /usr/include/Eigen
-sudo ln -sf /usr/include/eigen3/unsupported /usr/include/unsupported
+git clone [email protected]:mp3guy/ElasticFusion.git
+cd ElasticFusion/
+git submodule update --init
+cd third-party/OpenNI2/
+make -j8
+cd ../Pangolin/
+mkdir build
+cd build
+cmake .. -DEIGEN_INCLUDE_DIR=$HOME/ElasticFusion/third-party/Eigen/ -DBUILD_PANGOLIN_PYTHON=false
+make -j8
+cd ../../..
+mkdir build
+cd build/
+cmake ..
 ```
 
-***invalid use of incomplete type ‘const struct Eigen ...***
-
-Pangolin must be installed AFTER all the other libraries to make use of optional dependencies.
-
-***GLSL 3.30 is not supported. Supported versions are 1.10, 1.20, 1.30, 1.00 ES and 3.00 ES***
-
-Make sure you are running ElasticFusion on your nVidia GPU. In particular, if you have an Optimus GPU
-- If you use Prime, follow instructions [here](http://askubuntu.com/questions/661922/how-am-i-supposed-to-use-nvidia-prime)
-- If you use Bumblebee, remember to run as `optirun ./ElasticFusion`
-
-# 4. How do I use it? #
-There are three subprojects in the repo:
+# 2. How do I use it? #
+There are two subprojects in the repo:
 
 * The *Core* is the main engine which builds into a shared library that you can link into other projects and treat like an API. 
-* The *GUI* is the graphical interface used to run the system on either live sensor data or a logged data file. 
-* The *GPUTest* is a small benchmarking program you can use to tune the CUDA kernel launch parameters used in the main engine. 
+* The *Tools* where the graphical interface used to run the system on either live sensor data or a logged data file lives.
 
-The GUI (*ElasticFusion*) can take a bunch of parameters when launching it from the command line. They are as follows:
+The executable (*ElasticFusion*) can take a bunch of parameters when launching it from the command line. They are as follows:
 
 * *-cal <calibration>* : Loads a camera calibration file specified as *fx fy cx cy*.
 * *-l <logfile>* : Processes the specified .klg log file.
@@ -111,21 +71,15 @@ The GUI (*ElasticFusion*) can take a bunch of parameters when launching it from
 
 Essentially by default *./ElasticFusion* will try run off an attached ASUS sensor live. You can provide a .klg log file instead with the -l parameter. You can capture .klg format logs using either [Logger1](https://github.com/mp3guy/Logger1) or [Logger2](https://github.com/mp3guy/Logger2). 
 
-# 5. How do I just use the Core API? #
+# 3. How do I just use the Core API? #
 The libefusion.so shared library which gets built by the Core is what you want to link against.
 
-An example of this can be seen in the GUI code. Essentially all you need to do is utilise the provided Findefusion.cmake file in GUI/src and include the following in your CMakeLists.txt file:
-
-    find_package(efusion REQUIRED)
-    include_directories(${EFUSION_INCLUDE_DIR})
-    target_link_libraries(MyProject ${EFUSION_LIBRARY})
-
 To then use the Core API, make sure to include the header file in your source file:
 ```cpp
     #include <ElasticFusion.h>
 ```
 
-Initialise the static configuration parameters once somewhere at the start of your program (this [smells](http://en.wikipedia.org/wiki/Code_smell), but whatever):
+Initialise the static configuration parameters once somewhere at the start of your program:
 ```cpp
     Resolution::getInstance(640, 480);
     Intrinsics::getInstance(528, 528, 320, 240);
@@ -145,20 +99,20 @@ Make an ElasticFusion object and start using it:
     eFusion.processFrame(rgb, depth, timestamp, currentPose, weightMultiplier);
 ```
 
-See the source code of MainController.cpp in the GUI source to see more usage.
+See the source code of MainController.cpp to see more usage.
 
-# 6. Datasets #
+# 4. Datasets #
 
 We have provided a sample dataset which you can run easily with ElasticFusion for download [here](http://www.doc.ic.ac.uk/~sleutene/datasets/elasticfusion/dyson_lab.klg). Launch it as follows:
 
 ```bash
 ./ElasticFusion -l dyson_lab.klg
 ```
 
-# 7. License #
+# 5. License #
 ElasticFusion is freely available for non-commercial use only.  Full terms and conditions which govern its use are detailed [here](http://www.imperial.ac.uk/dyson-robotics-lab/downloads/elastic-fusion/elastic-fusion-license/) and in the LICENSE.txt file.
 
-# 8. FAQ #
+# 6. FAQ #
 ***What are the hardware requirements?***
 
 A [very fast nVidia GPU (3.5TFLOPS+)](https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units#GeForce_900_Series), and a fast CPU (something like an i7). If you want to use a non-nVidia GPU you can rewrite the tracking code or substitute it with something else, as the rest of the pipeline is actually written in the OpenGL Shading Language. 
@@ -169,7 +123,7 @@ Download [Stopwatch](https://github.com/mp3guy/Stopwatch) and run *StopwatchView
 
 ***I ran a large dataset and got assert(graph.size() / 16 < MAX_NODES) failed***
 
-Currently there's a limit on the number of nodes in the deformation graph down to lazy coding (using a really wide texture instead of a proper 2D one). So we're bound by the maximum dimension of a texture, which is 16384 on modern cards/OpenGL. Either fix the code so this isn't a problem any more, or increase the modulo factor in *Shaders/sample.geom*. 
+Currently there's a limit on the number of nodes in the deformation graph down to lazy coding (using a really wide texture instead of a proper 2D one). So we're bound by the maximum dimension of a texture, which is 16384 on modern cards/OpenGL. Either fix the code so this isn't a problem any more, or increase the modulo factor in *Core/Shaders/sample.geom*. 
 
 ***I have a nice new laptop with a good GPU but it's still slow***