SemArt dataset is a corpus with 21,384 samples that provides artistic comments along with fine-art paintings and their attributes for studying semantic art understanding.

Text2Art Challenge

Text2Art challenge, based on text-image retrieval, is used to evaluate the performance of semantic art understanding, whereby given an artistic text, a relevant image is found, and vice versa.

Human evaluation

For a given artistic comment and attributes, standard human evaluators (i.e. not art experts) were asked to choose the most appropriate painting from a pool of ten painting images. We evaluate the task in two different levels: easy and difficult.

• In the easy level, images shown for a given text are chosen randomly from all the painting images in test set.

• In the difficult level, ithe ten images shown for each comment share the same metadata field type.

Examples

Citation

@InProceedings{Garcia2018How,
   author    = {Noa Garcia and George Vogiatzis},
   title     = {How to Read Paintings: Semantic Art Understanding with Multi-Modal Retrieval},
   booktitle = {Proceedings of the European Conference in Computer Vision Workshops},
   year      = {2018},
}