janusch
janusch
I removed the pruning and use relocation. It seems to give better results. Here is one of the eval videos at 30k steps. I need to run the benchmarks :)...
Thank you for your suggestions @liruilong940607 I still play a little bit with the parameters and settings before I start cleaning it up. If the measure are not somewhat flawed...
Update: The last commit gives me PSNR: 27.941, SSIM: 0.8705, LPIPS: 0.074 Time: 0.012s/image Number of GS: 1213319
Here are my results. I think I overfitted a bit on the garden scene. What is the base line? According to [nerfbaselines](https://nerfbaselines.github.io/) it is sota (psnr) on the following scenes:...
Here is the latest state of this draft pr | Scene | PSNR | SSIM | LPIPS | Memory (GB) | Num GS (anchors) | |:--------|--------:|-------:|--------:|--------------:|-------------------:| | garden | 27.9297...
@ichsan2895 thx for testing. I think that confirms what is to be expected. Awesome!
> BTW, Can it save to ply and view it using supersplat or mkkellogg gaussian viewer? @ichsan2895 There is a way to do that. I thought gsplat is creating ply...
I know how to do it. Is it like a big feature request? I can code it up in open3D. I just want to get this here done for a...
I think that's good for review now.
@ichsan2895 No. I can't test it properly as I don't have a multi gpu setup. Either I remove it to not make it crash or I implement it blindly. @earth-bass...