Datasets:
Formats:
parquet
Size:
10K - 100K
metadata
dataset_info:
features:
- name: image
dtype: image
- name: conditioning_image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 15288004025.986
num_examples: 91521
download_size: 14120432425
dataset_size: 15288004025.986
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
task_categories:
- keypoint-detection
- image-to-image
- pose-estimation