qianqiuzzz
qianqiuzzz
Nice work! I want to know how to monitor the memory-use, I have done experiments on 3dgs and compGS, but I don't know how to compare their memory usage.
When I tried your method of processing camera parameters in the actorshq dataset and applied the processed extrinsic and intrinsic matrices to the original 3DGS, the results were very blurry....
When I use a dataset with no background, gaussians begins to become scattered, they appear in the background position after 7k iterations, causing black areas to appear in the mesh....
 Thanks for your wonderful work. When I try to run the example, this problem occurs and it seems always occur at Step 15000.
How to implement 2D hash encoder？ Do you have any plans to implement this? So it can work on kplanes
About the experiment results on dna_rendering, the gt images is raw formats with background , masked, or croped? I would like to know the details of your processing of dna-rendering
Can I get your optimized camera parameters of DNA-rendering datasets?