Surface between 2 polylines and calculating eroded volume

[nellett]

Hi all,
I am trying to calculate an eroded volume from a scoured debris flow channel. I have a dense post-erosion point cloud from SfM, but no pre-existing topography. Thus far, I have managed to get volumes from short (~20m) segments with the following steps: Create polylines along each bank where eroded/not eroded meet (the difference is based on color and form), extract profiles for those polylines, next create a mesh between those two polyline profiles, next sample points on the mesh. I think the image displays what I mean. From here, I used the Compute 2.5D volume tool, leaving the empty cells empty on both before and after, and compared the results with a few different grid steps. I am satisfied with the resulting volume.
However, I am having trouble expanding this approach to the full coverage I have of the eroded channel (~1km total) because I cannot accurately place polylines for the banks at the full scale, and I cannot merge separate polyline segments after creating them at appropriate scales.
Is there a recommended solution or work process? Does my setup make sense? Is there another completely different procedure to use? I would prefer to avoid doing the entire 1km of eroded volume in 20m segments of meshes between polylines.
One last issue- when exporting a geotiff from the rasterize tool and opening in ArcMap I can define projection and have the x,y location correct. However, the elevation values seem to carry over from being recentered in CloudCompare, rather than in the original coordinate system. In effect, after defining projection in Arc the raster is in the correct x,y locations, but elevation values are 1700m too low (1700m from global shift when points were originally brought into CloudCompare).
Thanks in advance for any inputs,
Nick

[daniel]

Not sure how you could improve your workflow. Have you tried to use the 'Trace Polyline tool' to delineate the borders of the channel, and then the 'Edit > Mesh > Surface between two polylines' tool?

Otherwise, regarding your shift issue, it may be a bug of CloudCompare (we probably forgot to apply the shift along Z when generating rasters!). Meanwhile, you should simply set the Global Shift along Z to 0 (when the Global Shift dialog appears).

[nellett]

I did try the trace polyline tool, but didnt have great luck with oversampling and getting the polyline to follow the cloud well. I'll give it another shot though. Thanks!
Nick