Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
16,950 datasets
Karine-Huang developed T2I-CompBench in 2023 to provide a standardized framework for evaluating compositionality in text-to-image models. The benchmark includes curated text prompts and evaluation metrics published in NeurIPS 2023 and TPAMI to assess how models handle complex textual instructions.
1460 coded political actors across 290 presidential administrations in 20 Latin American countries. Scott Mainwaring commissioned archival research reports between 2008 and 2013 to measure normative regime preferences and policy radicalism. The data collection covers the period from 1944 to 2010, with Argentina and El Salvador extending back to 1916 and 1927.
STIPLAR is a real-world scene text image dataset containing Korean, Arabic, and Japanese text image pairs. The data was collected from MLT-2019 and web sources and is designed for fine-tuning the STELLAR model on low-resource languages. It was created by yongchoooon and last updated on Hugging Face in November 2025.
The Yiddish Synthetic Pangoline Dataset is a collection of synthetic Yiddish document images generated using a custom Pangoline text-to-image synthesis tool. It contains high-quality synthetic Yiddish text rendered as images, along with corresponding ground truth text and ALTO-XML layout annotations. The dataset was created by author johnlockejrr and was last updated on 2025-11-02.
A qualitative study deposited in October 2025 examines how search systems impact systematic searching. Data were collected from interviews with twelve systematic searchers and analyzed using reflexive thematic analysis. The dataset includes deidentified interview transcripts, recruitment materials, and coded themes for two planned publications.
Vietnamese Handwriting Ocr is a dataset for optical character recognition tasks, hosted on the Hugging Face platform by user manhha2502. The dataset was last updated on December 16, 2025. Its specific contents, such as the number of samples or annotation format, are not detailed in the available metadata.
Intelligent Interaction Agent Dataset V0.1 is a large-scale, multi-modal dataset designed for building AI assistants. The dataset, created by deepgo and last updated on 2025-11-03, supports tasks like vehicle interaction recognition, multi-turn dialogue, and emotion-aware agent development.
Fantastic Beasts is a dataset collected for the NeurIPS 2023 paper 'AttrSeg: Open-Vocabulary Semantic Segmentation via Attribute Decomposition-Aggregation'. It was created by author chaofanma and last updated on Hugging Face in October 2025. The dataset is designed to address the lack of rare or obscure vocabulary in existing segmentation benchmarks.
753 substation images with polygon mask annotations across 18 object categories, plus three additional subsets for transmission line classification and detection. The collection features drone-captured imagery specifically targeting power system infrastructure for computer vision applications.
Infinity-Doc-55K contains 55,000 real-world and synthetic scanned documents for full-text parsing, published by infly in 2025. The collection features layout variations and structural annotations across financial, medical, and academic report domains.
GRAID BDD100K is a dataset of structured question-answer pairs generated from object detection annotations in driving scenes. The dataset was created by author kd7 using the GRAID framework and was last updated on Hugging Face in October 2025. It is designed to test aspects of object localization, visual reasoning, and spatial reasoning.
EgoExoBench is a benchmark designed to evaluate cross-perspective understanding in multimodal large models. It contains synchronized and asynchronous egocentric and exocentric video pairs with multiple-choice questions. The dataset was authored by Heleun and last updated on November 3, 2025.
A dataset of PDF documents annotated for OCR classification tasks, published by HuggingFaceFW. It contains binary labels (OCR/NOCR) and file size information for each PDF. The dataset was last updated on October 20, 2025.
A dataset of CT scans with dense segmentation annotations for 14 anatomical targets, including adrenal glands, colon, duodenum, esophagus, gallbladder, kidneys, liver, lungs, pancreas, small bowel, spleen, stomach, trachea, and bladder. The dataset is hosted on Hugging Face by author Angelou0516 and was last updated on October 30, 2025. Data is provided in the NIfTI (.nii.gz) format.
A protected microdata file from Statistics Netherlands (CBS) provides information on the relationship between individuals and the labor market. The survey covers persons aged 15 and older in the Netherlands, excluding the institutionalized population, and links personal characteristics to their current or future labor market position. This version contains revised data for 2003-2012, aligned with later survey years and includes new weighting variables to match official CBS tables.
GRAID NuImages is a question-answer dataset generated by the GRAID framework for enhancing spatial reasoning in Vision-Language Models. The dataset was created by author kd7 and is associated with a research paper and project page. It was last updated on October 29, 2025.
The TotalSegmentator Organs dataset contains CT scans with dense segmentation annotations for 14 anatomical structures. The dataset is provided by MedOtter and was last updated on October 30, 2025. The data format is NIfTI (.nii.gz).
Code-170k-shona is a dataset containing 176,999 programming conversations, originally sourced from glaiveai/glaive-code-assistant-v2 and translated into Shona. It was created by michsethowusu and last updated on October 30, 2025. The dataset aims to make coding education accessible to Shona speakers through multi-turn dialogues.
A subset of 100,000 anime character images from the Zerochan webdataset. The images are filtered to be non-monochrome and depict a single person, head, and face with one primary character. Annotator animetimm created this dataset, which was last updated on November 5, 2025.
OpenPecha's benchmark dataset evaluates Tibetan optical character recognition models. It includes diverse scripts, writing styles, and print methods to enable testing across multiple domains. The dataset was last updated on October 30, 2025.