Skip to content

Chinese City Parking Dataset (CCPD) 2019

Data Downloading

The CCPD can be downloaded from this link using either the Google or BaiduYun drive links. This dataset is divided into 3 sets: train, val, test. Labels for each set can be found under the splits directory of the dataset. CCPD-Green dataset is already separated into different folders and thus does not require labels.

The annotations for each image are embedded into the filename of the image. The format is described on their official website here.

After downloading the dataset, the directory structure should be like as follows: 

CCPD2019
  |--- ccpd_base
  |    |--- <image_name>.jpg
  |    |--- <image_name>.jpg
  |    |--- ...
  |--- ccpd_blur
  |    |--- <image_name>.jpg
  |    |--- <image_name>.jpg
  |    |--- ...
  |--- ...
  |--- ...
  |--- ...
  |--- splits

Data Preparation

For Detection Task

To prepare the data for text detection, you can run the following commands:

python tools/dataset_converters/convert.py \
    --dataset_name ccpd --task det \
    --image_dir path/to/CCPD2019/ccpd_base \
    --label_dir path/to/CCPD2019/splits/train.txt \
    --output_path path/to/CCPD2019/det_gt.txt

label_dir is not required for CCPD-Green dataset.

The generated standard annotation file det_gt.txt will now be placed under the folder CCPD2019/.

Back to dataset converters