This project focused on image classification, using the MAMe dataset - a collection distinguished by its high resolution and variable shape attributes, comprising thousands of artworks from three museums-.
The classification task involved differentiating between 29 artistic mediums. Through experiments, we evaluated the impact of high resolution images, variable shape inputs, and their combination. Results demonstrated enhanced performance with high resolution images and variable shapes.
Elena Alegret
Below is a list of the files included in this project, along with a brief description of each:
- Code: A file containing all the codes used to obtain the results.
- Results: A file containing all the results obtained from the executions organized by an identity number.
It is important to remarck that data is obtained by MAMe oficial website.