Skip to content

Latest commit

 

History

History
82 lines (72 loc) · 3.33 KB

README.md

File metadata and controls

82 lines (72 loc) · 3.33 KB
Raw

Preparing Few-Shot Segmentation Datasets

Download following datasets:

1. COCO-20i

Download COCO2014 train/val images and annotations:

wget http://images.cocodataset.org/zips/train2014.zip
wget http://images.cocodataset.org/zips/val2014.zip
wget http://images.cocodataset.org/annotations/annotations_trainval2014.zip

Download COCO2014 train/val annotations from this Google Drive: train2014.zip, val2014.zip. (and locate both train2014/ and val2014/ under annotations/ directory).

2. FSS-1000

Download FSS-1000 images and annotations from this Google Drive.

3. LVIS-92i

Download COCO2017 train/val images:

wget http://images.cocodataset.org/zips/train2017.zip
wget http://images.cocodataset.org/zips/val2017.zip

Download LVIS-92i extended mask annotations from our Google Drive: lvis.zip.

4. PACO-Part

Download COCO2017 train/val images:

wget http://images.cocodataset.org/zips/train2017.zip
wget http://images.cocodataset.org/zips/val2017.zip

Download PACO-Part extended mask annotations from our Google Drive: paco.zip.

5. Pascal-Part

Download VOC2010 train/val images:

wget http://roozbehm.info/pascal-parts/trainval.tar.gz
wget http://host.robots.ox.ac.uk/pascal/VOC/voc2010/VOCtrainval_03-May-2010.tar

Download Pascal-Part extended mask annotations from our Google Drive: pascal.zip.

Create a directory 'datasets' for the above datasets and appropriately place each dataset to have following directory structure:

datasets/
├── COCO2014/           
│   ├── annotations/
│   │   ├── train2014/
│   │   └── val2014/
│   ├── train2014/
│   ├── val2014/
│   └── splits
│   │   ├── trn/
│   │   └── val/
├── FSS-1000/
│   ├── data/
│   │   ├── ab_wheel/
│   │   ├── ...
│   │   └── zucchini/
│   └── splits/   
│   │   ├── test.text
│   │   ├── trn.txt
│   │   └── val.txt
├── LVIS/
│   ├── coco/
│   │   ├── train2017/
│   │   └── val2017/
│   ├── lvis_train.pkl
│   └── lvis_val.pkl
├── PACO-Part/
│   ├── coco/
│   │   ├── train2017/
│   │   └── val2017/
│   ├── paco/
│   │   ├── paco_part_train.pkl
│   │   └── paco_part_val.pkl
├── Pascal-Part/  
│   ├── VOCdevkit/
│   │   ├── VOC2010/
│   │   │   ├── Annotations_Part_json_merged_part_classes/
│   │   │   ├── JPEGImages/
│   │   │   └── all_obj_part_to_image.json