skywater_seg / data /README.md
Realcat's picture
Upload folder using huggingface_hub
ec20751 verified
|
Raw
History Blame Contribute Delete
1.38 kB

ADE20K Sky-Water-Person Split Files

These split files list ADE20K images that contain at least one of the target classes (sky, water, person).

Format

Each line is a path relative to the ADE20K image root (ADEChallengeData2016/images/):

validation/ADE_val_00000001.jpg
training/ADE_train_00000001.jpg

Expected Dataset Layout

ADEChallengeData2016/
β”œβ”€β”€ images/
β”‚   β”œβ”€β”€ training/
β”‚   β”‚   β”œβ”€β”€ ADE_train_00000001.jpg
β”‚   β”‚   └── ...
β”‚   └── validation/
β”‚       β”œβ”€β”€ ADE_val_00000001.jpg
β”‚       └── ...
└── annotations/
    β”œβ”€β”€ training/
    β”‚   β”œβ”€β”€ ADE_train_00000001.png
    β”‚   └── ...
    └── validation/
        β”œβ”€β”€ ADE_val_00000001.png
        └── ...

Class Mapping

ADE20K raw labels β†’ skywater 4-class scheme:

ADE20K ID ADE20K Name Target Class
3 sky 1 (Sky)
13 person 3 (Person)
22 water 2 (Water)
27 sea 2 (Water)
61 lake 2 (Water)
105 pond 2 (Water)
110 fountain 2 (Water)
114 waterfall 2 (Water)
129 river 2 (Water)
(all others) β€” 0 (Background)

Obtaining ADE20K

Download from: https://groups.csail.mit.edu/vision/datasets/ADE20K/

Use the ADEChallengeData2016 subset (training + validation).