emiliofidalgo/bimos

Binary descriptor-based Image Mosaicing

bimos's Issues

Question on Transform types

Hello Emilio;

There are 4 types declared in Transform.h , but only two have been implemented in HomographyEstimator.cpp
The two missing ones are PERSPECTIVE and ROTATIONAL.
Is that right?

ros kinetic

Dear Emilio, could you please kinetic the package...? ya estas tardando xaval en kinetizar ese pedasso de paquete.

Tu compañero, Paco Grande.

Hi Do you have any idea to make the result picture orthographic?

If I got the GPS info of each image, can I get an orthophoto from the result? Like a tile on a map?

OpenCV dependency for both obindex and bimos

Which version of OpenCV do both obindex and bimos depend on?

Error in compile

Running command: "make cmake_check_build_system" in "/home/lizecai/catkin_ws/build"

Running command: "make -j1 -l1" in "/home/lizecai/catkin_ws/build"

[ 6%] Built target binaryindex
[ 10%] Built target test_bindex
[ 13%] Building CXX object bimos/CMakeFiles/bimos_convfactors.dir/src/util/Params.cpp.o
In file included from /home/lizecai/catkin_ws/src/bimos/src/util/Params.cpp:21:0:
/home/lizecai/catkin_ws/src/bimos/include/bimos/util/Params.h:28:31: fatal error: bimos/BimosConfig.h: No such file or directory
#include <bimos/BimosConfig.h>
^
compilation terminated.
make[2]: *** [bimos/CMakeFiles/bimos_convfactors.dir/src/util/Params.cpp.o] Error 1
make[1]: *** [bimos/CMakeFiles/bimos_convfactors.dir/all] Error 2
make: *** [all] Error 2
Invoking "make -j1 -l1" failed
lizecai@ubuntu:~/catkin_ws$

Could not find ROS master. Either start a 'roscore' or abort loading the plugin.

Hi
Thanks for your shared code. I did all steps, exactly same as explained here. with the same version of Ubuntu and ROS. I did test obindex as well. But after all steps, when I run rqt and import bimos it says:
Could not find ROS master. Either start a 'roscore' or abort loading the plugin.
Could you please help me about that.
Best,
Babaie

Kinetic branch on Ubuntu 18.04 and using ROS-melodic

What parameters did you use to generate the sample Valldemossa result ?
I have run this source on the dataset and the result looks blurred.

blending takes a lot of time

The result shows that blending costs a lot more time than optimization and feature extraction and matching. What is the main cause of this?