Claude Monet

We apply archetypal analysis to a set of 1339 artworks by Claude Monet obtained from WikiArt. Since this collection is small compared to the full GanGogh collection, we choose \(k=32\) archetypes.

Archetype Visualizations

The figures below show all 32 archetypes.

/images/neurips18/monet/archetypes/wikiart_512_cov01234_gram_appendmean_triul_svd_full_std_4K_dimnorm_samplenorm_p32_split_00_04.jpg/images/neurips18/monet/archetypes/wikiart_512_cov01234_gram_appendmean_triul_svd_full_std_4K_dimnorm_samplenorm_p32_split_04_08.jpg/images/neurips18/monet/archetypes/wikiart_512_cov01234_gram_appendmean_triul_svd_full_std_4K_dimnorm_samplenorm_p32_split_08_12.jpg/images/neurips18/monet/archetypes/wikiart_512_cov01234_gram_appendmean_triul_svd_full_std_4K_dimnorm_samplenorm_p32_split_12_16.jpg/images/neurips18/monet/archetypes/wikiart_512_cov01234_gram_appendmean_triul_svd_full_std_4K_dimnorm_samplenorm_p32_split_16_20.jpg/images/neurips18/monet/archetypes/wikiart_512_cov01234_gram_appendmean_triul_svd_full_std_4K_dimnorm_samplenorm_p32_split_20_24.jpg/images/neurips18/monet/archetypes/wikiart_512_cov01234_gram_appendmean_triul_svd_full_std_4K_dimnorm_samplenorm_p32_split_24_28.jpg/images/neurips18/monet/archetypes/wikiart_512_cov01234_gram_appendmean_triul_svd_full_std_4K_dimnorm_samplenorm_p32_split_28_32.jpg

Style Enhancement

We enhance the strongest (left side) and second strongest (right side) found in the image relative to the rest of the decomposition.

/images/neurips18/monet/linear_enhancement/enhance_strongest_0018.jpg/images/neurips18/monet/linear_enhancement/enhance_strongest_0186.jpg/images/neurips18/monet/linear_enhancement/enhance_strongest_0294.jpg/images/neurips18/monet/linear_enhancement/enhance_strongest_0354.jpg/images/neurips18/monet/linear_enhancement/enhance_strongest_0498.jpg/images/neurips18/monet/linear_enhancement/enhance_strongest_0636.jpg

Free Style Manipulation

Interpolations between two archetypes on a grid. The archetypes are freely chosen, they are not necessarily present in the image decompositions before stylization.

/images/neurips18/monet/grids/freestyle_18.jpg/images/neurips18/monet/grids/freestyle_1242.jpg/images/neurips18/monet/grids/freestyle_1254.jpg/images/neurips18/monet/grids/freestyle_1272.jpg

Image Decompositions

Archetypal analysis allows to decompose an image into its contributing styles. Below are some examples of these decompositions. The contributions of the archetypal styles are visualized in three ways:

  • as a texture synthesized using the archetypal style (left column)
  • as stylization of the image in questions, using a unit vector for each contributing style (top row)
  • as a sum of the (strongest three) contributing images
/images/neurips18/monet/image_decomps/decomp_0012.jpg/images/neurips18/monet/image_decomps/decomp_0060.jpg/images/neurips18/monet/image_decomps/decomp_1320.jpg/images/neurips18/monet/image_decomps/decomp_1218.jpg/images/neurips18/monet/image_decomps/decomp_1170.jpg/images/neurips18/monet/image_decomps/decomp_1098.jpg/images/neurips18/monet/image_decomps/decomp_0990.jpg/images/neurips18/monet/image_decomps/decomp_0840.jpg/images/neurips18/monet/image_decomps/decomp_0756.jpg/images/neurips18/monet/image_decomps/decomp_0660.jpg