CGL-Dataset / README.md
shunk031's picture
Update CGLDataset dataset card
8762b7a verified
|
Raw
History Blame Contribute Delete
6.43 kB
metadata
annotations_creators:
  - crowdsourced
language:
  - zh
language_creators:
  - found
license:
  - cc-by-nc-sa-4.0
multilinguality:
  - monolingual
pretty_name: CGL-Dataset
size_categories:
  - 10K<n<100K
source_datasets:
  - original
tags:
  - graphic-design
  - poster
  - layout-generation
task_categories:
  - image-to-image
task_ids: []
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
      - split: test
        path: data/test-*
  - config_name: ralf-style
    data_files:
      - split: train
        path: ralf-style/train-*
      - split: validation
        path: ralf-style/validation-*
      - split: test
        path: ralf-style/test-*
      - split: no_annotation
        path: ralf-style/no_annotation-*
dataset_info:
  - config_name: default
    features:
      - name: image_id
        dtype: int64
      - name: file_name
        dtype: string
      - name: width
        dtype: int64
      - name: height
        dtype: int64
      - name: image
        dtype: image
      - name: annotations
        sequence:
          - name: area
            dtype: int64
          - name: bbox
            sequence: int64
          - name: category
            struct:
              - name: category_id
                dtype: int64
              - name: name
                dtype:
                  class_label:
                    names:
                      '0': logo
                      '1': text
                      '2': underlay
                      '3': embellishment
                      '4': highlighted text
              - name: supercategory
                dtype: string
    splits:
      - name: train
        num_bytes: 7727076720.09
        num_examples: 54546
      - name: validation
        num_bytes: 824988413.326
        num_examples: 6002
      - name: test
        num_bytes: 448856950
        num_examples: 1000
    download_size: 8848246626
    dataset_size: 9000922083.416
  - config_name: ralf-style
    features:
      - name: image_id
        dtype: int64
      - name: file_name
        dtype: string
      - name: width
        dtype: int64
      - name: height
        dtype: int64
      - name: original_poster
        dtype: image
      - name: inpainted_poster
        dtype: image
      - name: saliency_map
        dtype: image
      - name: saliency_map_sub
        dtype: image
      - name: annotations
        sequence:
          - name: area
            dtype: int64
          - name: bbox
            sequence: int64
          - name: category
            struct:
              - name: category_id
                dtype: int64
              - name: name
                dtype:
                  class_label:
                    names:
                      '0': logo
                      '1': text
                      '2': underlay
                      '3': embellishment
                      '4': highlighted text
              - name: supercategory
                dtype: string
    splits:
      - name: train
        num_bytes: 29834119281.261364
        num_examples: 48438
      - name: validation
        num_bytes: 3722970297.954319
        num_examples: 6055
      - name: test
        num_bytes: 3701864874.9093184
        num_examples: 6055
      - name: no_annotation
        num_bytes: 448869325
        num_examples: 1000
    download_size: 37543869068
    dataset_size: 37707823779.125

Dataset Card for CGL-Dataset

CI Sync HF

Dataset Description

Dataset Summary

CGL-Dataset is a poster layout dataset released with Composition-aware Graphic Layout GAN for Visual-Textual Presentation Designs. The paper studies layout generation for a given image, emphasizing that both global semantics and spatial image composition affect where graphic elements should be placed. The original dataset contains 60,548 advertising posters with annotated layout information.

Supported Tasks and Leaderboards

The dataset supports poster layout generation and layout-conditioned graphic design modeling. No public leaderboard is bundled with this Hugging Face dataset.

Languages

Poster text is primarily Chinese (zh).

Dataset Structure

Data Fields

The default config contains image_id, file_name, width, height, image, and COCO-style annotations.

The ralf-style config provides original posters, inpainted posters, saliency maps, and annotations for layout-generation pipelines.

Data Splits

Config Split Rows
default train 54,546
default validation 6,002
default test 1,000
ralf-style train 48,438
ralf-style validation 6,055
ralf-style test 6,055
ralf-style no_annotation 1,000

Dataset Creation

The dataset was created for Composition-aware Graphic Layout GAN research. It provides visual element categories and positions for poster layout generation, enabling models to synthesize text and decorative layouts conditioned on image content rather than using template-only rules.

Considerations for Using the Data

The data focuses on advertising poster layouts and may reflect the visual conventions of the source domain.

Additional Information

Licensing Information

The dataset card uses the CC BY-NC-SA 4.0 metadata from the local loader.

Citation Information

@inproceedings{ijcai2022p692,
  title     = {Composition-aware Graphic Layout GAN for Visual-Textual Presentation Designs},
  author    = {Zhou, Min and Xu, Chenchen and Ma, Ye and Ge, Tiezheng and Jiang, Yuning and Xu, Weiwei},
  booktitle = {Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, {IJCAI-22}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  pages     = {4995--5001},
  year      = {2022},
  doi       = {10.24963/ijcai.2022/692},
  url       = {https://doi.org/10.24963/ijcai.2022/692}
}

Contributions

Thanks to minzhouGithub for creating the original dataset.