User Tools

Site Tools


tamiwiki:projects:scanning-tami

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
tamiwiki:projects:scanning-tami [2025/02/18 12:57] wissotskytamiwiki:projects:scanning-tami [2025/02/24 13:24] (current) wissotsky
Line 1: Line 1:
 <WRAP center round todo 60%> <WRAP center round todo 60%>
-Very WIP page(as of 18th February 2025), you can help by asking questions in the telegram group or in-person on Mondays+Very WIP page(as of 24th February 2025), you can help by asking questions in the telegram group or in-person on Mondays
 </WRAP> </WRAP>
  
Line 23: Line 23:
  
 After that I switched to superpoint for the feature detection and lightglue for feature matching which seems to be a fairly popular combo currently. After that I switched to superpoint for the feature detection and lightglue for feature matching which seems to be a fairly popular combo currently.
 +
 +{{tamiwiki:projects:animation_tamiscan_superpoint_lightglue.gif}}
  
 First I matched every frame with its 30 nearest frames(by time). First I matched every frame with its 30 nearest frames(by time).
Line 58: Line 60:
  
 I was able smooth the depth over by rasterizing the median depth(according to RaDe-GS) I was able smooth the depth over by rasterizing the median depth(according to RaDe-GS)
 +
 +{{tamiwiki:projects:combined_depths_animation.gif}}
  
 But I still had some outliers in the depth data. But I still had some outliers in the depth data.
Line 70: Line 74:
  
 I was able to fit a kernel density estimate to the depth data and then I manually found cutoff value where if the density after the global peak becomes lower it means that were past the primary peak any depth beyond that is an outlier. I was able to fit a kernel density estimate to the depth data and then I manually found cutoff value where if the density after the global peak becomes lower it means that were past the primary peak any depth beyond that is an outlier.
 +
 +{{tamiwiki:projects:threshold_animation.gif}}
  
 After removing the depth outliers I was able to get much cleaner results After removing the depth outliers I was able to get much cleaner results
  
-#TODO tsdf integration+To get a mesh from the depth images I used TSDF integration, I used the VoxelBlockGrid implementation from open3d. 
 + 
 +But the gpu vram wasnt enough for me to extract mesh detail down to 1mm. And running the integration purely on cpu was too slow. 
 + 
 +So I ended up computing the tsdf volume in batches on the gpu and them merging them onto a uniform voxel grid on the cpu, where there was overlap between the grids I used trilinear interpolation.
  
-mesh compresson+#TODO mesh compresson
  
 voxelization voxelization
tamiwiki/projects/scanning-tami.1739876273.txt.gz · Last modified: 2025/02/18 12:57 by wissotsky