Skip to content

s3dis datasets need preprocessing? #672

Description

@yfq512

Checklist

My Question

I am trying to reproduce the training on the S3DIS dataset.
The dataset dir is

datasets/
└── S3DIS
    └── Stanford3dDataset_v1.2_Aligned_Version
        ├── Area_1
        │   ├── pantry_1
        │   │   └── Annotations
        │   └── WC_1
        │       └── Annotations
        ├── Area_2
        │   ├── pantry_1
        │   │   └── Annotations
        │   └── WC_1
        │       └── Annotations
        ├── Area_3
        │   ├── pantry_1
        │   │   └── Annotations
        │   └── WC_1
        │       └── Annotations
        ├── Area_4
        │   ├── pantry_1
        │   │   └── Annotations
        │   └── WC_1
        │       └── Annotations
        ├── Area_5
        │   ├── pantry_1
        │   │   └── Annotations
        │   └── WC_1
        │       └── Annotations
        ├── Area_6
        │   ├── pantry_1
        │   │   └── Annotations
        │   └── WC_1
        │       └── Annotations
        └── original_pkl

Do I need to manually perform data preprocessing operations?

Activity

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

    questionFurther information is requested

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions