Results format

[XilSLAM]

Hi,

I'm trying to use your algorithm to extract 3D line segments from my monocular dataset but I don't understand the format you used in your .obj segments files. (v x y z and l a b for example in line_segments_clustered_incr.obj)
As it looks like a point cloud format, when I try to open it with meshlab it just shows me a point cloud and no segments...

Do you know how I could print the segments instead of the points in meshlab and how to read the v x y z/l a b information not as point coordinates but as segments information ?

Thanks.

[shidahe]

Hi,

Thanks for trying it out.

In the .obj file, each point is defined by its 3D coordinate (x, y, z). The points are indexed by their order. Then each line segment is therefore defined by the indices of two points (a, b).

I was using MeshLab to read the .obj file, so MeshLab should be able to show the line segment in this format. There are some viewing options I have to select (some buttons to click) in the top tool bar of Meshlab, in order to let it show the line segments properly. Sorry it's been a lone time so I forget what exactly those options are, but it should be easy to figure it out by trying out different display configurations.