Skip to content

The size of the data obtained by training with OpenSplat is half the size of the Python version #101

Closed

Description

@Sqiuln

OpenSplat is really great! But I've encountered some strange phenomena when using it. When I trained on the official dataset for 3D Gaussian Splatting, I found that the size of the .ply files generated by the training is half the size of the Python version of 3DGS.This could potentially lead to differences in the rendering results between OpenSplat and 3D Gaussian Splatting.
Can you tell me what the reason for this phenomenon is? I set the number of training iterations to 30,000.
OpenSplat 3DgaussianSplatting
playroom 487,752kb 838,642kb
drjohnson 256,947kb 571,773kb
train 175,448kb 258,677kb
truck 282,510kb 631,085kb

Activity

  1. locked and limited conversation to collaborators on Jun 11, 2024
  2. converted this issue into a discussion #102 on Jun 11, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions