The ArtBench Dataset: Benchmarking Generative Models with Artworks (2022-06-22T00:00:00.000000Z)

TL;DR

This work introduces ArtBench-10, the first class-balanced, high-quality, cleanly annotated, and standardized dataset for benchmarking artwork generation, and provides three versions of the dataset, formatted in a way that is easy to be incorporated by popular machine learning frameworks.

Abstract

We introduce ArtBench-10, the first class-balanced, high-quality, cleanly annotated, and standardized dataset for benchmarking artwork generation. It comprises 60,000 images of artwork from 10 distinctive artistic styles, with 5,000 training images and 1,000 testing images per style. ArtBench-10 has several advantages over previous artwork datasets. Firstly, it is class-balanced while most previous artwork datasets suffer from the long tail class distributions. Secondly, the images are of high quality with clean annotations. Thirdly, ArtBench-10 is created with standardized data collection, annotation, filtering, and preprocessing procedures. We provide three versions of the dataset with different resolutions ($32\times32$, $256\times256$, and original image size), formatted in a way that is easy to be incorporated by popular machine learning frameworks. We also conduct extensive benchmarking experiments using representative image synthesis models with ArtBench-10 and present in-depth analysis. The dataset is available at https://github.com/liaopeiyuan/artbench under a Fair Use license.

Authors

K. Keutzer

18 papers

Xiuyu Li

3 papers

Peiyuan Liao

1 papers

TL;DR

Abstract

Authors

References61 items

Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding

Hierarchical Text-Conditional Image Generation with CLIP Latents

Pseudo Numerical Methods for Diffusion Models on Manifolds

The Met Dataset: Instance-level Recognition for Artworks

StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasets

High-Resolution Image Synthesis with Latent Diffusion Models

GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models

Projected GANs Converge Faster

Rebooting ACGAN: Auxiliary Classifier GANs with Stable Training

Alias-Free Generative Adversarial Networks

Diffusion Models Beat GANs on Image Synthesis

Improved Denoising Diffusion Probabilistic Models

End-to-End Chinese Landscape Painting Creation Using Generative Adversarial Networks

Denoising Diffusion Implicit Models

NVAE: A Deep Hierarchical Variational Autoencoder

Denoising Diffusion Probabilistic Models

Differentiable Augmentation for Data-Efficient GAN Training

Training Generative Adversarial Networks with Limited Data

Analyzing and Improving the Image Quality of StyleGAN

Consistency Regularization for Generative Adversarial Networks

Generative Modeling by Estimating Gradients of the Data Distribution

Improved Precision and Recall Metric for Assessing Generative Models

A Style-Based Generator Architecture for Generative Adversarial Networks

Large Scale GAN Training for High Fidelity Natural Image Synthesis

How to Read Paintings: Semantic Art Understanding with Multi-Modal Retrieval

Museum Exhibit Identification Challenge for the Supervised Domain Adaptation and Beyond

Datasheets for datasets

Demystifying MMD GANs

DeepArt: Learning Joint Representations of Visual Arts

GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium

BAM! The Behance Artistic Media Dataset for Recognition Beyond Photography

ArtGAN: Artwork synthesis with conditional categorical GANs

Improved Techniques for Training GANs

If Art Education Then Critical Digital Making: Computational Thinking and Creative Code

LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop

Deep Unsupervised Learning using Nonequilibrium Thermodynamics

Deep Learning Face Attributes in the Wild

The Rijksmuseum Challenge: Museum-Centered Visual Recognition

Recognizing Image Style

Artistic Image Classification: An Analysis on the PRINTART Database

An Analysis of Single-Layer Networks in Unsupervised Feature Learning

ImageNet: A large-scale hierarchical image database

Transforming Information Into Knowledge: How Computational Methods Reshape Art History

AUTO-ENCODING VARIATIONAL BAYES

NoisyArt: A Dataset for Webly-supervised Artwork Recognition

The imet collection 2019 challenge

GENERATIVE ADVERSARIAL NETS

Omniart: A large-scale artistic benchmark

At the Met

Unpaired image-to-image translation using cycle-consistent adversarial networks

The State of the Art: Object Retrieval in Paintings using Discriminative Regions

Looks like it

Learning Multiple Layers of Features from Tiny Images

Art, emergence, and the computational sublime

Gradient-based learning applied to document recognition

Edinburgh Research Explorer The PASCAL Visual Object Classes (VOC) Challenge

Imagemagick v6 examples – resampling by nicolas robidoux, Oct 2012

for visual classification and content-based retrieval of artistic content containing 112,039 artworks of ancient times, medieval ages and the late 1900s exhibited in the Rijksmuseum Amsterdam

It is collected from public knowledge bases like DBpedia at first, and then queried from Google Image Search and Flickr

) and contains 8,629 oil paintings of 10 classes (aeroplane, bird, boat, chair, cow, dining-table, dog, horse, sheep, train). The resolution of images is around 500 pixels in width

We mainly conduct our experiments with NVIDIA GPUs. See Table 5 for the detailed configuration

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names