Sign in to view source links and access this dataset
Description
790,000 Arabic and Chinese annotations extend the ArtEmis collection of 80,000 artworks and 450,000 English emotion labels. This dataset, created by youssef101, serves as a benchmark for evaluating AI systems on cultural and linguistic diversity. It was last updated on Hugging Face in July 2023.
Use Cases
Benchmarking cultural-transfer performance in AI systems based on the described multilingual annotations.
Training or evaluating multimodal models for emotion labeling based on the described artwork and caption data.
Studying cross-linguistic and cross-cultural differences in emotional responses to visual art based on the described Arabic and Chinese extensions.
Strengths
Extends a known dataset (ArtEmis) with 790,000 new annotations in Arabic and Chinese.
Builds upon a base of 80,000 artworks and 450,000 English emotion labels.
Explicitly designed for a specific research purpose: evaluating 'cultural-transfer' in AI.
Limitations
Column-level documentation is absent; field semantics must be inferred after download.
Row count for the full combined dataset is unknown, which may limit suitability assessment.
Freshness should be verified as the last update was 2023-07-23.
Provenance
Source
Extension of the ArtEmis dataset, which sourced 80,000 artworks from WikiArt.
Collection Method
Likely involves manual or semi-automated annotation to add emotion labels and captions.
Time Range
null
Freshness
Last updated 2023-07-23 16:21:23.
Geography
null
License is unknown; terms of use must be verified before application.