DATASET.md

August 17, 2021 ยท View on GitHub

DATASET

Our main 20 predicate category VG settings follows split from Limited-Labels.

Download:

  1. Download the VG images part1 (9 Gb) part2 (5 Gb). Extract these images to the file datasets/vg/VG_100K. If you want to use other directory, please link it in DATASETS['VG_stanford_filtered']['img_dir'] of maskrcnn_benchmark/config/paths_catelog.py.
  2. Download the 20 settings from GoogleDrive. Extract these files to datasets/ and forms the file structure datasets/vg/20.