Datasets:

Modalities:
Image
Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
License:
SDIP_dog / README.md
rmokady's picture
Update README.md
b9e1266 verified
metadata
license: apache-2.0
dataset_info:
  features:
    - name: image
      dtype: image
  splits:
    - name: train
      num_bytes: 15763120370.036
      num_examples: 126841
  download_size: 15856205293
  dataset_size: 15763120370.036
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

This repository contains Unofficial access for SDIP-dogs dataset

Official Repository, Project Page, Paper

Self-Distilled Internet Photos (SDIP) is a multi-domain image dataset. The dataset consists of Self-Distilled Flickr (SD-Flickr) and *Self-Distilled LSUN (SD-LSUN) that were crawled from Flickr and LSUN dataset, respectively, and then curated using the method described in our Self-Distilled StyleGAN paper:

Self-Distilled StyleGAN: Towards Generation from Internet Photos
Ron Mokady, Michal Yarom, Omer Tov, Oran Lang, Daniel Cohen-Or, Tali Dekel, Michal Irani, Inbar Mosseri

Overview

StyleGAN’s fascinating generative and editing abilities are limited to structurally aligned and well-curated datasets. It does not work well on raw datasets downloaded from the Internet. The SDIP domains presented here, which are StyleGAN-friendly, were automatically curated by our method from raw images collected from the Internet. The raw uncurated images in Self-Distilled Flicker (SD-Flickr) were first crawled from Flickr using a simple keyword (e.g. 'dog' or 'elephant').

The dataset in this page exhibits 4 domains: SD-Dogs (126K images) 1024X1024, SD-Elephants (39K images) 512X512, SD-Bicycles (96K images) 256X256, and SD-Horses (162K images) 256X256. Our curation process consists of a simple pre-processing step (off-the-shelf object detector to crop the main object and then rescale), followed by a sophisticated StyleGAN-friendly filtering step (which removes outlier images while maintaining dataset diversity). This results in a more coherent and clean dataset, which is suitable for training a StyleGAN2 generator (see more details in our paper).

The data itself is saved in a json format: for SD-Flickr we provide urls of the original images and bounding boxes used for cropping; for SD-LSUN we provide image identifiers with the bounding boxes. In addition to the SDIP dataset, we also provide weights of pre-trained StyleGAN2 models trained using each image domain presented in the paper.