Closed Saadi4469 closed 2 years ago
You cannot dynamically edit a 2D plot so what is proposed in lidR in 3D is not reproducible in 2D with regular plots.
Edit: Or a raster but in 3D. Why not.
Well actually it is already working. I'm closing
LASfile <- system.file("extdata", "MixedConifer.laz", package="lidR")
las = readLAS(LASfile)
chm = rasterize_canopy(las)
ttops = locate_trees(chm, manual())
If there was a way I know to get a faster rendering I swear I would already have put it in lidR :wink:
Oh cool thank you so much, I will give it a try tomorrow at work. However, I wish there was a way to fasten rgl 3D rendering for manual tree correction purposes.
Hope you had some fruitful discussion with the rgl
maintainer.
If you are referring to the person on stackoverflow I thnk they are not the rgl maintainer. And we did not discuss at all actually. But I'm working on improving the rgl
rendering for billion of points. I think I have the ideas clear but the implementation is hard.
I am so happy to hear that, thank you so much for your time and help. I did post the same question here, and the rgl
maintainer did modify some of the code to improve performance. Good luck!
He did point out that the performance has to do with the CPU and not the GPU when it comes to rendering points. Moreover, rgl
is single threaded and therefore does not use multiple thread, and the maintainer told me it's not possible to make rgl
use multiple threads.
Good day,
Based on this issue I would like to suggest an optional way to manually correct tree tops if possible by overlaying the points on an aerial basemap or a high resolution raster. This might increase the performance of the process. I am also open to other solutions as well.
Thank you