README.md

metadata

license: apache-2.0

Download following datasets:

1. PASCAL-5ⁱ

Download PASCAL VOC2012 devkit (train/val data):

wget http://host.robots.ox.ac.uk/pascal/VOC/voc2012/VOCtrainval_11-May-2012.tar

Download PASCAL VOC2012 SDS extended mask annotations.

2. 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. (and locate both train2014/ and val2014/ under annotations/ directory).

Create a directory '../dataset' for the above few-shot segmentation datasets and appropriately place each dataset to have following directory structure:

Datasets/
  ├── VOC2012/            # PASCAL VOC2012 devkit
  │   ├── Annotations/
  │   ├── ImageSets/
  │   ├── ...
  │   └── SegmentationClassAug/
  ├── COCO2014/           
  │   ├── annotations/
  │   │   ├── train2014/  # (dir.) training masks (from Google Drive) 
  │   │   ├── val2014/    # (dir.) validation masks (from Google Drive)
  │   │   └── ..some json files..
  │   ├── train2014/
  │   └── val2014/
  └── IC-VOS/
      ├── 0a43a414/
      │   ├── Annotations/
      │   └── JPEGImages/
      └── ...

Citation

@article{qi2025dc,
  title={DC-SAM: In-Context Segment Anything in Images and Videos via Dual Consistency},
  author={Qi, Mengshi and Zhu, Pengfei and Ma, Huadong and Qi, Lu and Li, Xiangtai and Yang, Ming-Hsuan},
  journal={arXiv preprint arXiv:2504.12080},
  year={2025}
}