### scripts for processing the datasets structure for vidor dataset ``` vidor ├── train │   └── video │   ├── 0000 │   ├── 0001 │   ├── 0002 ├── train_annotation │   └── training │   ├── 0000 │   ├── 0001 │   ├── 0002 │   ├── 0003 ├── validation │   └── video │   ├── 0001 │   ├── 0004 │   ├── 0005 ├── validation_annotation │   └── validation │   ├── 0001 │   ├── 0004 ``` structure for vitstg ``` VidSTG-Dataset ├── annotations │   ├── read_annotation.py │   ├── test_annotations.json │   ├── test_files.json │   ├── train_annotations.json │   ├── train_files.json │   ├── val_annotations.json │   └── val_files.json ├── example.jpg └── README.md ```