aislabunimi / ROSE2

Robust structure identification and room segmentation of cluttered indoor environments from occupancy grid maps
http://wiki.ros.org/rose2
GNU General Public License v3.0
16 stars 2 forks source link

Questions about how to get the room segmentation result #2

Closed B0GGY closed 1 year ago

B0GGY commented 1 year ago

Hi, thanks for your work. I am wondering how could I save the room segmentation result like Fig5 from the paper 'Robust structure identification and room segmentation of cluttered indoor environments from occupancy grid map' as a png image.

goldleaf3i commented 1 year ago

Hi, if you are working with images as input, and you are not using the method online as embedded with ROS, you can use the old version of the code (here https://github.com/goldleaf3i/declutter-reconstruct) that saves the image. Also, you can find there how to save such a file. there is a function that does what you need.

The code of the two repo is basically the same, this one is a little more commented, updated, and integrated with ros. I can look into it later this week to give more details if you need them

B0GGY commented 1 year ago

I tried the code from https://github.com/goldleaf3i/declutter-reconstruct, but it says ModuleNotFoundError: No module named 'fft_structure_extraction'. So I found one from https://github.com/tkucner/rose/blob/master/fft_structure_extraction.py. However, it seems that some methods and attributes are missing, like show() and generate_initial_hypothesis_simple().

Nisici commented 1 year ago

Hello, you can get the room segmentation result as a png image launching voronoi.launch file, it will store outputs in the folder src/rose_v2_repo/map_processing. It's going to be slower than the method using ROSE.launch because it's doing postprocessing using voronoi method that finds more rooms.

B0GGY commented 1 year ago

Thanks a lot, it works!