The intersection of computer vision and art history has grown rapidly, enabling tasks such as artist attribution, style classification, and digital cataloging. Early benchmarks like the ArtClass v1 dataset provided a foundational 50-class artist classification task [1]. However, real-world art collections present more nuanced challenges: an artwork may belong to multiple overlapping styles (e.g., “Impressionism” and “Landscape”), span multiple temporal categories, or include ambiguous attributions.
Table 1: Comparison of generative performance on artistic benchmarks. artclass v2
find an active link to the site? AI can make mistakes, so double-check responses Copy Creating a public link... You can now share this thread with others Good response Bad response 4 sites proudparrot2/artclass-v2: Official repository for Art Class Web, a site ... Table_title: proudparrot2/artclass-v2 Table_content: header: | Name | Name | Last commit message | Last commit date | row: | Name: GitHub proudparrot2/artclass-v2 - Codesandbox Edit the code to make changes and see it instantly in the preview. Explore this online proudparrot2/artclass-v2 sandbox and experi... CodeSandbox Official repository for Art Class v2, the best site for unblocked games ... Table_title: art-class/v2 Table_content: header: | Name | Name | row: | Name: Latest commit History 18 Commits 18 Commits | Name: ... GitHub Collection of Educational Resources | PDF - Scribd You might also like * Educational Resource Links Compilation. No ratings yet. Educational Resource Links Compilation. 4 pages. * E... Scribd 4 sites proudparrot2/artclass-v2: Official repository for Art Class Web, a site ... Table_title: proudparrot2/artclass-v2 Table_content: header: | Name | Name | Last commit message | Last commit date | row: | Name: GitHub proudparrot2/artclass-v2 - Codesandbox Edit the code to make changes and see it instantly in the preview. Explore this online proudparrot2/artclass-v2 sandbox and experi... CodeSandbox Official repository for Art Class v2, the best site for unblocked games ... Table_title: art-class/v2 Table_content: header: | Name | Name | row: | Name: Latest commit History 18 Commits 18 Commits | Name: ... GitHub Show all The intersection of computer vision and art history
Utilizes proxy technologies or "about:blank" cloaking techniques to help users bypass standard content filters. Version History and Legacy Table 1: Comparison of generative performance on artistic
Fine-grained visual classification (FGVC) of artwork is challenging due to high intra-class variance, subtle inter-class differences, and domain-specific attributes (e.g., brushwork, palette, era). We introduce , a new benchmark dataset consisting of 120,000 labeled artwork images spanning 150 artist styles, 12 historical periods, and 8 medium types (oil, watercolor, etc.). Unlike its predecessor, ArtClass v2 provides multi-label annotations (style + period + subject matter) and is designed to handle real-world art collection scenarios with class imbalance and partial labels. We evaluate 10 state-of-the-art FGVC architectures (e.g., DenseNet, Vision Transformers, MLP-Mixers) and show that even top models achieve only 68.3% top-1 accuracy, leaving significant room for improvement. ArtClass v2 is publicly available to spur research in computational art history and digital humanities.
| Model | Top-1 (std) | Top-1 (hard) | mAP (multi-label) | |------------------|-------------|--------------|-------------------| | ResNet-50 | 62.1% | 41.3% | 0.534 | | ViT-B/16 | 68.3% | 48.7% | 0.612 | | Swin-T | 67.5% | 47.9% | 0.604 | | ConvNeXt-B | | 49.2% | 0.615 |
ArtClass v2 utilizes a U-Net backbone operating in a latent space of $8 \times 8$ downsampling. The core innovation lies in the fine-tuning of the attention layers on the ArtClass-Corpus.