Plugins

Supervisely/ Pascal VOC

Import Updated An hour ago Free
To upload Pascal VOC Semnatic Segmentation dataset
Free Signup

Pascal VOC Import

Usage steps:

1) Download Pascal VOC dataset from official site. (Or use direct Link)

2) Unpack archive

3) Directory structure have to be the following:

.
├── ImageSets
│   └── Segmentation
│       ├── train.txt
│       ├── trainval.txt
│       └── val.txt
├── JPEGImages
│   ├── 2007_000032.jpg
│   ├── 2007_000033.jpg
│   ├── ...
├── SegmentationClass
│   ├── 2007_000032.png
│   ├── 2007_000033.png
│   ├── ...
└── SegmentationObject
    ├── 2007_000032.png
    ├── 2007_000033.png
    ├── ...

4) Open Supervisely import page. Choose Pascal VOC import plugin.

5) Select all subdirectories (ImageSets, JPEGImages, SegmentationClass, SegmentationClass) and drag and drop them to browser. Wait a little bit.

6) Define new project name and click on START IMPORT button.

7) After import task finish, you can view project and see follow datasets: train, trainval, val.

![](https://imgur.com/37jUZZ1.jpg)

8) Datasets samples contains images and instance segmentation annotations. See few examples:

![](https://i.imgur.com/hJ93iv3.jpg)

![](https://i.imgur.com/UVqlFlp.jpg)

9) On Statistics project tab you can get more information, for example - class distribution:

![](https://i.imgur.com/BS79Qr2.png)

Notes:

  • If you will drag and drop parent directory instead of its content, import will crash.
  • Supervisely import documentation.

Docker Image

supervisely/import-pascal:latest

More Info

ID
69827
Version ID
84490
First released
7 months ago
Last updated
An hour ago

Owner

s