Research map / Label-efficient and robust learning / Recognition and long-tail
A Deep Learning Approach to Clustering Visual Arts
Giovanna Castellano, Gennaro Vessio
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use DenseNet121 pretrained features plus deep embedded clustering instead of k-means on the autoencoder embedding | DELIUS · WikiArt 8 style periods | clustering accuracy: 0.25 -> 0.28 | Improved |
| joint deep embedded clustering instead of k-means on the pretrained autoencoder embedding | DELIUS · WikiArt 10 genres | clustering accuracy: 0.43 -> 0.47 | Improved |
| use pretrained DenseNet features before embedding instead of raw images (DCEC) | DELIUS · WikiArt overall | silhouette and Calinski-Harabasz | Improved |
| add deep embedding and clustering layer instead of PCA+k-means or AE+k-means | DELIUS · WikiArt overall | silhouette and Calinski-Harabasz | Improved |
| use UMAP instead of t-SNE for visualization | DELIUS · WikiArt overall | cluster visual separation | Improved |
| use DenseNet121 instead of ResNet50 as backbone (supervised style classification) (ablation) | backbone · WikiArt 8 styles | overfitting behavior | Improved |
Closest papers by meaning
- A Learned Representation For Artistic Style (ICLR 2017 · similarity 0.86)
- Learning to Discover Novel Visual Categories via Deep Transfer Clustering (ICCV 2019 · similarity 0.86)
- Proxy Learning of Visual Concepts of Fine Art Paintings from Styles through Language Models (AAAI 2022 · similarity 0.85)
- Deep Transformation-Invariant Clustering (NeurIPS 2020 · similarity 0.85)
- Deep Clustering for Unsupervised Learning of Visual Features (ECCV 2018 · similarity 0.84)
- The Shape of Art History in the Eyes of the Machine (AAAI 2018 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
