You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
thanks for the really nice repo and code.
i have tried dataset and seen the result of the clustering.
however as i am using jetson orin 64gb developer kits. the fps of the clustering was bit slow.
is there a reason? and if it is possible
the thing that i wanna accomplishing is using the clustering algorithms alone with fastlio2
not making the TSDF mapping. is it possible?
thanks for the reading and have a great day!
(and i just figured out that dynablox is included in the nvblox, i will try that one and tell you how it works.
again have a great day!)
The text was updated successfully, but these errors were encountered:
Thanks for your interest! The motion cue leveraged by Dynablox to detect moving objects is free space - which requires building a TSDF. Clustering only happens to group the detected dynamic points but won't do much without the motion detection first.
You can play with larger voxel sizes or lower integration ranges if high framerate is really important t you, or use the more GPU-optimized parallelized version in nvblox. Hope this helps!
thanks for the really nice repo and code.
i have tried dataset and seen the result of the clustering.
however as i am using jetson orin 64gb developer kits. the fps of the clustering was bit slow.
is there a reason? and if it is possible
the thing that i wanna accomplishing is using the clustering algorithms alone with fastlio2
not making the TSDF mapping. is it possible?
thanks for the reading and have a great day!
(and i just figured out that dynablox is included in the nvblox, i will try that one and tell you how it works.
again have a great day!)
The text was updated successfully, but these errors were encountered: