Unpaired Van Gogh Paintings And Photographs For Image Translation

Comprising unpaired images of Van Gogh paintings and photographs, originally part of the CycleGAN benchmark. It is used for training unpaired image-to-image translation models. The dataset is associated with the 2017 research paper on Cycle-Consistent Adversarial Networks.

Use Cases

• Train CycleGAN models for unpaired image-to-image translation between Van Gogh painting and photograph domains.
• Benchmark style transfer algorithms using the distinct artistic style of Van Gogh paintings.
• Evaluate generative adversarial networks on the task of translating real-world photographs to artistic styles.

Strengths

• Part of the established CycleGAN benchmark suite for image translation research.
• Associated with a seminal 2017 computer vision paper (arXiv:1703.10593).

Limitations

• Dataset size and specific image counts are not provided in the input.
• The data is unpaired, which may limit supervised learning approaches.

Provenance

Source

Originally hosted by the Berkeley AI Research (BAIR) Laboratory, associated with the CycleGAN project.

Collection Method

Curated collection of Van Gogh paintings and photographs.

Freshness

Last updated on the platform in April 2022.