updated readme

README.md

@@ -85,6 +85,14 @@ vikit is a catkin project, therefore, download it into your catkin workspace sou
     cd catkin_ws/src
     git clone https://github.com/uzh-rpg/rpg_vikit.git
 
+#### ROS dependencies
+
+Sometimes `cmake-modules` is missing (required for including Eigen in ROS Indigo):
+
+    sudo apt-get install ros-hydro-cmake-modules
+
+(replace `hydro` with your distribution).
+
 #### SVO
 
 Now we are ready to build SVO.
@@ -95,12 +103,6 @@ Clone it into your catkin workspace
 
 If you installed g2o then set `HAVE_G2O` in `svo/CMakeLists.txt` to TRUE.
 
-Install missing ros dependencies. Sometimes `cmake-modules` is missing (required for including Eigen in ROS Indigo):
-
-    sudo apt-get install ros-hydro-cmake-modules
-
-(replace `hydro` with your distribution).
-
 Finally, build:
 
     catkin_make
@@ -160,6 +162,14 @@ Parameter Settings
 
 A description of all parameters which can be set via the launchfile is provided in `svo/include/config.h`. The default parameters can be viewed in `svo/src/config.cpp`. Moreover, some additional parameters (mainly rostopic names etc.) are read from the ros parameter server in `svo_ros/vo_node.cpp`.
 
+#### Forward motions
+
+The current keyframe selection criterion is designed for downward looking cameras. This is one reason why SVO does not work well for forward motions (e.g., to be used on a car).
+
+#### Image resolution
+
+The current parameters are tuned for WVGA resolution. If you use a higher resolution camera, then the pyramid levels should be increased as well as the reprojection error thresholds.
+
 
 Contributing
 ------------

svo/test/README.md

@@ -1,11 +1,4 @@
 Download Synthetic Test Data:
 -----------------------------
 
-Downloads and extracts the test data in the folder `data`:
+TODO
-
-    cd rpg_slam/svo/test/data
-    wget http://rpg.ifi.uzh.ch/datasets/flying_room_1_rig_1.tar.gz -O - | tar -xz
-
-If you want to store the data in another location (for instance, if you maintain a special folder with all datasets), then you can move the dataset there and set the environment variable `TEST_DATA_DIR` in your `.bashrc` like this:
-
-    export SVO_DATASET_DIR=${HOME}/Datasets

svo_analysis/README.md

@@ -16,6 +16,7 @@ Source your new `.bashrc` script, go to the new dataset folder and download the
     cd ${SVO_DATASET_DIR}
     wget http://rpg.ifi.uzh.ch/datasets/flying_room_1_rig_1.tar.gz -O - | tar -xz
 
+TODO: replace this dataset with new ones.
 
 #### Run Benchmark