Updated to V0.3beta

- RGB-D compatibility, the RGB-D examples had been adapted to the new version.

- Kitti and TUM dataset compatibility, these examples had been adapted to the new version.

- ROS compatibility, It had been updated the old references in the code to work with this version.

- Config file parser, the YAML file contains the session configuration, a wrong parametrization may break the execution without any information to solve it. This version parses the file to read all the fields and give a proper answer if one of the fields have been wrong deffined or don't exist.

- Fixed minor bugs.

Author: richard-elvira

.gitignore

@@ -46,3 +46,4 @@ cmake_modules/
 cmake-build-debug/
 
 *.pyc
+*.osa

Changelog.md

@@ -0,0 +1,26 @@
+# ORB-SLAM3
+Details of changes between the different versions.
+
+### V0.3: Beta version, 4 Sep 2020
+
+- RGB-D compatibility, the RGB-D examples had been adapted to the new version.
+
+- Kitti and TUM dataset compatibility, these examples had been adapted to the new version.
+
+- ROS compatibility, It had been updated the old references in the code to work with this version.
+
+- Config file parser, the YAML file contains the session configuration, a wrong parametrization may break the execution without any information to solve it. This version parses the file to read all the fields and give a proper answer if one of the fields have been wrong deffined or don't exist.
+
+- Fixed minor bugs.
+
+
+### V0.2: Beta version, 7 Aug 2020
+Initial release. It has these capabilities:
+
+- Multiple-Maps capabilities, it is able to handle multiple maps in the same session and merge them when a common area is detected with a seamless fussion.
+
+- Inertial sensor, the IMU initialization takes a 2 seconds to achieve a scale error less than 5\% and it is reffined in the next 10 seconds until is around 1\%. Inertial measures are integrated at frame rate to estimate the scale, gravity and velocity in order to improve the visual features detection and make the system robust to temporal occlusions.
+
+- Fisheye sensor, the fisheye sensors are now fully supported in monocular and stereo. 
+
+

Dependencies.md

@@ -1,5 +1,5 @@
 ##List of Known Dependencies
-###ORB-SLAM3 version 0.2
+###ORB-SLAM3 version 0.3
 
 In this document we list all the pieces of code included  by ORB-SLAM3 and linked libraries which are not property of the authors of ORB-SLAM3.
 

Examples/Monocular/EuRoC.yaml

@@ -16,6 +16,9 @@ Camera.k2: 0.07395907
 Camera.p1: 0.00019359
 Camera.p2: 1.76187114e-05
 
+Camera.width: 752
+Camera.height: 480
+
 # Camera frames per second 
 Camera.fps: 20.0
 

Examples/Monocular/KITTI00-02.yaml

@@ -22,6 +22,10 @@ Camera.fps: 10.0
 # Color order of the images (0: BGR, 1: RGB. It is ignored if images are grayscale)
 Camera.RGB: 1
 
+# Camera resolution
+Camera.width: 1241
+Camera.height: 376
+
 #--------------------------------------------------------------------------------------------
 # ORB Parameters
 #--------------------------------------------------------------------------------------------

Examples/Monocular/KITTI03.yaml

@@ -22,6 +22,10 @@ Camera.fps: 10.0
 # Color order of the images (0: BGR, 1: RGB. It is ignored if images are grayscale)
 Camera.RGB: 1
 
+# Camera resolution
+Camera.width: 1241
+Camera.height: 376
+
 #--------------------------------------------------------------------------------------------
 # ORB Parameters
 #--------------------------------------------------------------------------------------------

Examples/Monocular/KITTI04-12.yaml

@@ -22,6 +22,10 @@ Camera.fps: 10.0
 # Color order of the images (0: BGR, 1: RGB. It is ignored if images are grayscale)
 Camera.RGB: 1
 
+# Camera resolution
+Camera.width: 1241
+Camera.height: 376
+
 #--------------------------------------------------------------------------------------------
 # ORB Parameters
 #--------------------------------------------------------------------------------------------

Examples/Monocular/TUM1.yaml

@@ -23,6 +23,10 @@ Camera.fps: 30.0
 # Color order of the images (0: BGR, 1: RGB. It is ignored if images are grayscale)
 Camera.RGB: 1
 
+# Camera resolution
+Camera.width: 640
+Camera.height: 480
+
 #--------------------------------------------------------------------------------------------
 # ORB Parameters
 #--------------------------------------------------------------------------------------------

Examples/Monocular/TUM2.yaml

@@ -23,6 +23,10 @@ Camera.fps: 30.0
 # Color order of the images (0: BGR, 1: RGB. It is ignored if images are grayscale)
 Camera.RGB: 1
 
+# Camera resolution
+Camera.width: 640
+Camera.height: 480
+
 #--------------------------------------------------------------------------------------------
 # ORB Parameters
 #--------------------------------------------------------------------------------------------

Examples/Monocular/TUM3.yaml

@@ -22,6 +22,10 @@ Camera.fps: 30.0
 # Color order of the images (0: BGR, 1: RGB. It is ignored if images are grayscale)
 Camera.RGB: 1
 
+# Camera resolution
+Camera.width: 640
+Camera.height: 480
+
 #--------------------------------------------------------------------------------------------
 # ORB Parameters
 #--------------------------------------------------------------------------------------------

Examples/RGB-D/TUM2.yaml

@@ -35,6 +35,9 @@ ThDepth: 40.0
 # Deptmap values factor 
 DepthMapFactor: 5208.0
 
+Camera.width: 640
+Camera.height: 480
+
 #--------------------------------------------------------------------------------------------
 # ORB Parameters
 #--------------------------------------------------------------------------------------------

Examples/RGB-D/TUM3.yaml

@@ -34,6 +34,9 @@ ThDepth: 40.0
 # Deptmap values factor
 DepthMapFactor: 5000.0
 
+Camera.width: 640
+Camera.height: 480
+
 #--------------------------------------------------------------------------------------------
 # ORB Parameters
 #--------------------------------------------------------------------------------------------

Examples/Stereo-Inertial/EuRoC.yaml

@@ -29,7 +29,7 @@ Camera.bf: 47.90639384423901
 Camera.RGB: 1
 
 # Close/Far threshold. Baseline times.
-ThDepth: 35 # 35
+ThDepth: 35.0 # 35
 
 # Transformation from camera 0 to body-frame (imu)
 Tbc: !!opencv-matrix

Examples/Stereo-Inertial/TUM_512.yaml

@@ -39,8 +39,11 @@ Tlr: !!opencv-matrix
             -0.000656143613644,  -0.046896036240590,   0.998899560146304,   0.001015350132563]
 
 # Lapping area between images
-Lapping.left: 0
-Lapping.right: 511
+Camera.lappingBegin: 0
+Camera.lappingEnd: 511
+
+Camera2.lappingBegin: 0
+Camera2.lappingEnd: 511
 
 # Camera resolution
 Camera.width: 512
@@ -53,7 +56,7 @@ Camera.fps: 20.0
 Camera.RGB: 1
 
 # Close/Far threshold. Baseline times.
-ThDepth: 40
+ThDepth: 40.0
 Camera.bf: 19.3079
 
 

Examples/Stereo-Inertial/TUM_512_outdoors.yaml

@@ -39,8 +39,11 @@ Tlr: !!opencv-matrix
             -0.000656143613644,  -0.046896036240590,   0.998899560146304,   0.001015350132563]
 
 # Lapping area between images
-Lapping.left: 0
-Lapping.right: 511
+Camera.lappingBegin: 0
+Camera.lappingEnd: 511
+
+Camera2.lappingBegin: 0
+Camera2.lappingEnd: 511
 
 # Camera resolution
 Camera.width: 512
@@ -53,7 +56,7 @@ Camera.fps: 20.0
 Camera.RGB: 1
 
 # Close/Far threshold. Baseline times.
-ThDepth: 40
+ThDepth: 40.0
 Camera.bf: 19.3079
 
 

Examples/Stereo/EuRoC.yaml

@@ -31,7 +31,7 @@ Camera.bf: 47.90639384423901
 Camera.RGB: 1
 
 # Close/Far threshold. Baseline times.
-ThDepth: 35
+ThDepth: 35.0
 
 #--------------------------------------------------------------------------------------------
 # Stereo Rectification. Only if you need to pre-rectify the images.

Examples/Stereo/KITTI00-02.yaml

@@ -31,7 +31,7 @@ Camera.bf: 386.1448
 Camera.RGB: 1
 
 # Close/Far threshold. Baseline times.
-ThDepth: 35
+ThDepth: 35.0
 
 #--------------------------------------------------------------------------------------------
 # ORB Parameters

Examples/Stereo/KITTI03.yaml

@@ -16,8 +16,6 @@ Camera.k2: 0.0
 Camera.p1: 0.0
 Camera.p2: 0.0
 
-Camera.bFishEye: 0
-
 Camera.width: 1241
 Camera.height: 376
 
@@ -31,7 +29,7 @@ Camera.bf: 387.5744
 Camera.RGB: 1
 
 # Close/Far threshold. Baseline times.
-ThDepth: 40
+ThDepth: 40.0
 
 #--------------------------------------------------------------------------------------------
 # ORB Parameters

Examples/Stereo/KITTI04-12.yaml

@@ -31,7 +31,7 @@ Camera.bf: 379.8145
 Camera.RGB: 1
 
 # Close/Far threshold. Baseline times.
-ThDepth: 40
+ThDepth: 40.0
 
 #--------------------------------------------------------------------------------------------
 # ORB Parameters

Examples/Stereo/TUM_512.yaml

@@ -46,10 +46,16 @@ Tlr: !!opencv-matrix
             -0.000823363992158,   0.998899461915674,   0.046895490788700,   0.001946204678584,
             -0.000656143613644,  -0.046896036240590,   0.998899560146304,   0.001015350132563]
 
+# Camera resolution
+Camera.width: 512
+Camera.height: 512
+
 # Lapping area between images
-Lapping.left: 0
-Lapping.right: 511
+Camera.lappingBegin: 0
+Camera.lappingEnd: 511
 
+Camera2.lappingBegin: 0
+Camera2.lappingEnd: 511
 
 # Camera frames per second
 Camera.fps: 20.0
@@ -58,7 +64,7 @@ Camera.fps: 20.0
 Camera.RGB: 1
 
 # Close/Far threshold. Baseline times.
-ThDepth: 40
+ThDepth: 40.0
 
 Camera.bf: 19.3079
 

README.md

@@ -1,8 +1,10 @@
 # ORB-SLAM3
 
-### V0.3: Beta version, 7 Aug 2020
+### V0.3: Beta version, 4 Sep 2020
 **Authors:** Carlos Campos, Richard Elvira, Juan J. Gómez Rodríguez, [José M. M. Montiel](http://webdiis.unizar.es/~josemari/), [Juan D. Tardos](http://webdiis.unizar.es/~jdtardos/).
 
+The [Changelog](https://github.com/UZ-SLAMLab/ORB_SLAM3/Changelog.md) describes the features of each version.
+
 ORB-SLAM3 is the first real-time SLAM library able to perform **Visual, Visual-Inertial and Multi-Map SLAM** with **monocular, stereo and RGB-D** cameras, using **pin-hole and fisheye** lens models. In all sensor configurations, ORB-SLAM3 is as robust as the best systems available in the literature, and significantly more accurate. 
 
 We provide examples to run ORB-SLAM3 in the [EuRoC dataset](http://projects.asl.ethz.ch/datasets/doku.php?id=kmavvisualinertialdatasets) using stereo or monocular, with or without IMU, and in the [TUM-VI dataset](https://vision.in.tum.de/data/datasets/visual-inertial-dataset) using fisheye stereo or monocular, with or without IMU. Videos of some example executions can be found at [ORB-SLAM3 channel](https://www.youtube.com/channel/UCXVt-kXG6T95Z4tVaYlU80Q).

include/MapDrawer.h

@@ -47,6 +47,8 @@ class MapDrawer
 
 private:
 
+    bool ParseViewerParamFile(cv::FileStorage &fSettings);
+
     float mKeyFrameSize;
     float mKeyFrameLineWidth;
     float mGraphLineWidth;

include/Tracking.h

@@ -62,6 +62,11 @@ class Tracking
 
     ~Tracking();
 
+    // Parse the config file
+    bool ParseCamParamFile(cv::FileStorage &fSettings);
+    bool ParseORBParamFile(cv::FileStorage &fSettings);
+    bool ParseIMUParamFile(cv::FileStorage &fSettings);
+
     // Preprocess the input and call Track(). Extract features and performs stereo matching.
     cv::Mat GrabImageStereo(const cv::Mat &imRectLeft,const cv::Mat &imRectRight, const double &timestamp, string filename);
     cv::Mat GrabImageRGBD(const cv::Mat &imRGB,const cv::Mat &imD, const double &timestamp, string filename);

include/Viewer.h

@@ -61,6 +61,8 @@ class Viewer
     bool both;
 private:
 
+    bool ParseViewerParamFile(cv::FileStorage &fSettings);
+
     bool Stop();
 
     System* mpSystem;

src/Frame.cc

@@ -1047,8 +1047,8 @@ Frame::Frame(const cv::Mat &imLeft, const cv::Mat &imRight, const double &timeSt
     mvInvLevelSigma2 = mpORBextractorLeft->GetInverseScaleSigmaSquares();
 
     // ORB extraction
-    thread threadLeft(&Frame::ExtractORB,this,0,imLeft,0,511);
-    thread threadRight(&Frame::ExtractORB,this,1,imRight,0,511);
+    thread threadLeft(&Frame::ExtractORB,this,0,imLeft,static_cast<KannalaBrandt8*>(mpCamera)->mvLappingArea[0],static_cast<KannalaBrandt8*>(mpCamera)->mvLappingArea[1]);
+    thread threadRight(&Frame::ExtractORB,this,1,imRight,static_cast<KannalaBrandt8*>(mpCamera2)->mvLappingArea[0],static_cast<KannalaBrandt8*>(mpCamera2)->mvLappingArea[1]);
     threadLeft.join();
     threadRight.join();
     std::chrono::steady_clock::time_point t2 = std::chrono::steady_clock::now();