CoolFace
30 shown

datasets

Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.

modalityAllTextImageAudioTabularVideo
taskAllText generationQuestion answeringImage classificationSpeech
sizeAllUnder 1K1K – 10K100K – 1M1M – 10M
formatAllParquetJSONCSVWebDataset
sortTrendingMost likedMost usedRecently updatedNewest
Clear all
01echo840 /OCRBenchGithub|Paper OCRBench has been accepted by Science China Information Sciences. image1K<n<10K25 likes19k downloads2y agoHugging Face02getomni-ai /ocr-benchmark OmniAI OCR Benchmark A comprehensive benchmark that compares OCR and data extraction capabilities of different multimodal LLMs such as gpt-4o and gemini-2.0, evaluating both text and JSON extraction accuracy. Benchmark Results (Feb 2025) | Source Code image1K<n<10K73 likes3.1k downloads2y agoHugging Face03ling99 /OCRBench_v2image10K<n<100K20 likes2k downloads2y agoHugging Face04lmms-lab /OCRBench-v2image10K<n<100K12 likes869 downloads2y agoHugging Face05Reza2kn /persian-ocr-bench-submitted10-bbox-crops Persian OCR benchmark — selected submitted bbox crops This dataset contains the non-empty OCR bboxes from the ten explicitly selected submitted pages in persian_ocr_bench_bbox_review. Each row is one PNG crop. gold_text is the current editable OCR content from the live Argilla bbox field (content_text). Geometry is stored both as source page pixels and as percentages of the source page. The original record ID, external ID, bbox ID, source URL, and SHA-256 hashes are included for… See the full description on the dataset page: https://huggingface.co/datasets/Reza2kn/persian-ocr-bench-submitted10-bbox-crops.imageimage-to-textn<1K1 likes656 downloads28d agoHugging Face06ussooraj /OCR-bench-Malayalamimageimage-to-text1K<n<10K0 likes637 downloads2mo agoHugging Face07tadad /kat57-ocr-bench-500 Kat57 OCR outputs Raw outputs from 16 OCR models on the same deterministic 500-card sample of Lund University Library's Kat57 catalogue-card collection. Each model is stored as a separate dataset configuration. Every configuration retains the source card identifiers, image, PAGE XML reference transcription, model output, and inference metadata so the results can be rescored without rerunning inference. Source sample CER/WER results and limitations ocr-bench… See the full description on the dataset page: https://huggingface.co/datasets/tadad/kat57-ocr-bench-500.image1K<n<10K0 likes563 downloads22d agoHugging Face08QYWH /ocrbench_v2text10K<n<100K0 likes548 downloads1y agoHugging Face09ilsilfverskiold /ocr-benchmark OCR Benchmark — Documents The 93 document images and ground truth used by the ocr-benchmark harness. The benchmark code, the reference run results, and the full methodology live in the GitHub repo — this dataset is the document corpus only. Structure One train split, 93 rows, one row per document: Column Type Description image Image The document page (PNG/JPG) stem string Filename stem (e.g. invoice_000) tier string Difficulty: easy, medium, or hard… See the full description on the dataset page: https://huggingface.co/datasets/ilsilfverskiold/ocr-benchmark.imageimage-to-textn<1K0 likes490 downloads2mo agoHugging Face10sfd-anonymous /edgar-ocr-benchmark EDGAR-OCR Benchmark This repository contains the accepted EDGAR-OCR benchmark samples used to evaluate OCR-style transcription of complex SEC filing tables. Contents samples/<sample_id>/screenshot.png: rendered table image used as model input. samples/<sample_id>/synthetic_table.html: synthetic filing-style HTML table used to render the input. samples/<sample_id>/ground_truth_table.md: target table transcription. samples/<sample_id>/ground_truth_grid.json: parsed target… See the full description on the dataset page: https://huggingface.co/datasets/sfd-anonymous/edgar-ocr-benchmark.image-to-text0 likes452 downloads5mo agoHugging Face11himalaya-ai /nepalipixel-synthetic-ocr-benchmark NepaliPixel Benchmark Dataset Model Card Overview The output_benchmark directory contains a synthetic OCR benchmark dataset for Nepali (Devanagari) script generated using the Nepali Pixel pipeline. The dataset is designed for evaluation of OCR models with minimal augmentation noise and fully rendered pages. Data Samples: Approximately 15,000 image‑text pairs (generated with -n 15000). Granularity: Includes all five levels – word, sentence… See the full description on the dataset page: https://huggingface.co/datasets/himalaya-ai/nepalipixel-synthetic-ocr-benchmark.imageimage-to-text10K<n<100K5 likes421 downloads3mo agoHugging Face12bhumika-tewari-282006 /assistive-ocr-benchmark-results Assistive OCR — Benchmark Results Real, reproducible benchmark results for the assistive OCR wearable module (offline, multilingual — English, Bengali+English, Hindi+English). This repository is self-contained: it holds the results, the ground-truth manifest, and the 98 real images they were computed from, so it can be run and demoed directly with no other dataset needed. What's in this repository File What it is manual100_final.csv The 99-row… See the full description on the dataset page: https://huggingface.co/datasets/bhumika-tewari-282006/assistive-ocr-benchmark-results.imageimage-to-textn<1K0 likes351 downloads2mo agoHugging Face13meharuhanzz /OCR-Bench1000-Gujarati OCR-Bench1000-Gujarati 1000 synthetic printed-text line images with ground-truth transcriptions, sampled from a larger locally-held Gujarati OCR training corpus. This is a benchmark/sample release, not the full training set. Data fields Field Description file_name relative path to the image (images/...) text ground-truth transcription category gujarati_only / english_only / mixed / numeric_and_symbols length_bucket short / medium / long, by… See the full description on the dataset page: https://huggingface.co/datasets/meharuhanzz/OCR-Bench1000-Gujarati.imageimage-to-text1K<n<10K0 likes345 downloads11d agoHugging Face14morpheushoc /OCRBenchv2OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning https://github.com/Yuliang-Liu/MultimodalOCR https://arxiv.org/abs/2501.00321 from datasets import load_dataset repo_dir = 'morpheushoc/OCRBenchv2' # load all samples dataset = load_dataset(repo_dir, split='test') # 10k samples # load a subset (EN/CN samples) dataset = load_dataset(repo_dir, 'EN', split='test') # 7.4k samples dataset =… See the full description on the dataset page: https://huggingface.co/datasets/morpheushoc/OCRBenchv2.image10K<n<100K3 likes317 downloads2y agoHugging Face15mshojaei77 /persian-ocr-bench Persian OCR Evaluation Dataset This benchmark contains paired Persian document images and UTF-8 transcription targets for OCR evaluation. Each JSONL row references one image under bench_data/images/ and contains its transcription in the text field. Schema image: image path relative to the repository id: stable image identifier page: page number, currently 1 type: evaluation item type, currently transcription text: reference transcription language: fa checked:… See the full description on the dataset page: https://huggingface.co/datasets/mshojaei77/persian-ocr-bench.imagen<1K1 likes262 downloads2mo agoHugging Face16katarinagresova /Genomic_Benchmarks_human_ocr_ensembl Dataset Card for "Genomic_Benchmarks_human_ocr_ensembl" More Information needed text100K<n<1M0 likes259 downloads4y agoHugging Face17meharuhanzz /OCR-Bench1000-Hindi OCR-Bench1000-Hindi 1000 synthetic printed-text line images with ground-truth transcriptions, sampled from a larger locally-held Hindi OCR training corpus. This is a benchmark/sample release, not the full training set. Data fields Field Description file_name relative path to the image (images/...) text ground-truth transcription category hindi_only / english_only / mixed / numeric_and_symbols length_bucket short / medium / long, by character count… See the full description on the dataset page: https://huggingface.co/datasets/meharuhanzz/OCR-Bench1000-Hindi.imageimage-to-text1K<n<10K0 likes243 downloads11d agoHugging Face18meharuhanzz /OCR-Bench1000-Malayalam OCR-Bench1000-Malayalam 1000 synthetic printed-text line images with ground-truth transcriptions, sampled from a larger locally-held Malayalam OCR training corpus. This is a benchmark/sample release, not the full training set. Data fields Field Description file_name relative path to the image (images/...) text ground-truth transcription category malayalam_only / english_only / mixed / numeric_and_symbols length_bucket short / medium / long, by… See the full description on the dataset page: https://huggingface.co/datasets/meharuhanzz/OCR-Bench1000-Malayalam.imageimage-to-text1K<n<10K0 likes237 downloads11d agoHugging Face19openpecha /OCR-Tibetan_line_to_text_benchmark Tibetan OCR-line-to-text Benchmark Dataset This repository hosts a line-to-text benchmark dataset to evaluate and compare Tibetan OCR models. The dataset includes diverse scripts, writing styles, and print methods, enabling comprehensive testing across multiple domains. 💽 Datasets Overview Features: filename: Name of the file. label: Ground truth text. image_url: URL of the image. BDRC_work_id: BDRC scan id for specific works. char_len: Character count of the text. script:… See the full description on the dataset page: https://huggingface.co/datasets/openpecha/OCR-Tibetan_line_to_text_benchmark.image100K<n<1M4 likes235 downloads11mo agoHugging Face20meharuhanzz /OCR-Bench1000-Bengali OCR-Bench1000-Bengali 1000 synthetic printed-text line images with ground-truth transcriptions, sampled from a larger locally-held Bengali OCR training corpus. This is a benchmark/sample release, not the full training set. Data fields Field Description file_name relative path to the image (images/...) text ground-truth transcription category bengali_only / english_only / mixed / numeric_and_symbols length_bucket short / medium / long, by character… See the full description on the dataset page: https://huggingface.co/datasets/meharuhanzz/OCR-Bench1000-Bengali.imageimage-to-text1K<n<10K0 likes172 downloads11d agoHugging Face21meharuhanzz /OCR-Bench1000-Marathi OCR-Bench1000-Marathi 1000 synthetic printed-text line images with ground-truth transcriptions, sampled from a larger locally-held Marathi OCR training corpus. This is a benchmark/sample release, not the full training set. Data fields Field Description file_name relative path to the image (images/...) text ground-truth transcription category marathi_only / english_only / mixed / numeric_and_symbols length_bucket short / medium / long, by character… See the full description on the dataset page: https://huggingface.co/datasets/meharuhanzz/OCR-Bench1000-Marathi.imageimage-to-text1K<n<10K0 likes171 downloads11d agoHugging Face22BDRC /tibetan-ocr-benchmark BDRC Tibetan OCR Benchmark (open subset) A hand-transcribed benchmark for evaluating Tibetan OCR across writing styles and technologies. This is an open-access set of 472 page images with ground-truth transcriptions and per-page metadata (script, technology, legibility). Companion to the model BDRC/tibetan-ocr and the leaderboard (dozens of OCR systems scored on this benchmark). The transcriptions were produced by Dharmaduta. The images were selected through detailed research… See the full description on the dataset page: https://huggingface.co/datasets/BDRC/tibetan-ocr-benchmark.imageimage-to-textn<1K0 likes170 downloads1mo agoHugging Face23meharuhanzz /OCR-Bench1000-Tamil OCR-Bench1000-Tamil 1000 synthetic printed-text line images with ground-truth transcriptions, sampled from a larger locally-held Tamil OCR training corpus. This is a benchmark/sample release, not the full training set. Data fields Field Description file_name relative path to the image (images/...) text ground-truth transcription category tamil_only / english_only / mixed / numeric_and_symbols length_bucket short / medium / long, by character count… See the full description on the dataset page: https://huggingface.co/datasets/meharuhanzz/OCR-Bench1000-Tamil.imageimage-to-text1K<n<10K0 likes167 downloads11d agoHugging Face24meharuhanzz /OCR-Bench1000-Telugu OCR-Bench1000-Telugu 1000 synthetic printed-text line images with ground-truth transcriptions, sampled from a larger locally-held Telugu OCR training corpus. This is a benchmark/sample release, not the full training set. Data fields Field Description file_name relative path to the image (images/...) text ground-truth transcription category telugu_only / english_only / mixed / numeric_and_symbols length_bucket short / medium / long, by character… See the full description on the dataset page: https://huggingface.co/datasets/meharuhanzz/OCR-Bench1000-Telugu.imageimage-to-text1K<n<10K0 likes165 downloads11d agoHugging Face25tadad /kat57-ocr-bench-500-results Kat57 OCR benchmark — CER/WER Strict reference-based evaluation of 16 OCR models on a deterministic 500-card sample from Lund University Library's Kat57 catalogue-card collection. This result set contains only Character Error Rate (CER) and Word Error Rate (WER); it does not contain VLM judging or ELO ratings. The sample was drawn with seed 57 from tadad/kat57-ground-truth and is published as tadad/kat57-ground-truth-500. The OCR outputs are retained in… See the full description on the dataset page: https://huggingface.co/datasets/tadad/kat57-ocr-bench-500-results.tabular1K<n<10K0 likes164 downloads22d agoHugging Face26meharuhanzz /OCR-Bench1000-Kannada OCR-Bench1000-Kannada 1000 synthetic printed-text line images with ground-truth transcriptions, sampled from a larger locally-held Kannada OCR training corpus. This is a benchmark/sample release, not the full training set. Data fields Field Description file_name relative path to the image (images/...) text ground-truth transcription category kannada_only / english_only / mixed / numeric_and_symbols length_bucket short / medium / long, by character… See the full description on the dataset page: https://huggingface.co/datasets/meharuhanzz/OCR-Bench1000-Kannada.imageimage-to-text1K<n<10K0 likes163 downloads11d agoHugging Face27ONTHEIT /KDoc-OCRBench-V2 KDoc-OCRBench-V2 A Korean document OCR benchmark, built in the style of olmOCR-Bench. 849 single-page PDFs from Korean public-sector documents with 56,197 manually-reviewed unit tests across three categories. Quick summary 849 single-page PDFs. Every PDF has at least one verified test in some category (no PDF is empty). Category Tests Type PDFs covered text_present 6,290 sentence-level presence 780 / 849 tables 49,115 table cell + adjacency 760 / 849… See the full description on the dataset page: https://huggingface.co/datasets/ONTHEIT/KDoc-OCRBench-V2.documentimage-to-textn<1K0 likes161 downloads4mo agoHugging Face28Reza2kn /bangla-ocr-double-benchmark Bangla OCR Double Benchmark Two equally weighted, deterministic full-page Bangla handwriting robustness splits: bongabdo: 6,669 readability-preserving renderings balanced over all 111 Bongabdo pages. bn_htrd: 6,669 renderings balanced over all 75 actual files in the writer-separated BN-HTRd test split. These are explicitly compositional/augmentation robustness rows, not 13,338 independent writers or source documents. Every row exposes its source page ID, source SHA-256… See the full description on the dataset page: https://huggingface.co/datasets/Reza2kn/bangla-ocr-double-benchmark.imageimage-to-text10K<n<100K0 likes161 downloads2mo agoHugging Face29meharuhanzz /OCR-Bench1000-Punjabi OCR-Bench1000-Punjabi 1000 synthetic printed-text line images with ground-truth transcriptions, sampled from a larger locally-held Punjabi OCR training corpus. This is a benchmark/sample release, not the full training set. Data fields Field Description file_name relative path to the image (images/...) text ground-truth transcription category punjabi_only / english_only / mixed / numeric_and_symbols length_bucket short / medium / long, by character… See the full description on the dataset page: https://huggingface.co/datasets/meharuhanzz/OCR-Bench1000-Punjabi.imageimage-to-text1K<n<10K0 likes155 downloads11d agoHugging Face30meharuhanzz /OCR-Bench1000-Sanskrit OCR-Bench1000-Sanskrit 1000 synthetic printed-text line images with ground-truth transcriptions, sampled from a larger locally-held Sanskrit OCR training corpus. This is a benchmark/sample release, not the full training set. Data fields Field Description file_name relative path to the image (images/...) text ground-truth transcription category sanskrit_only / english_only / mixed / numeric_and_symbols length_bucket short / medium / long, by… See the full description on the dataset page: https://huggingface.co/datasets/meharuhanzz/OCR-Bench1000-Sanskrit.imageimage-to-text1K<n<10K0 likes152 downloads11d agoHugging Face

Listings come live from the Hugging Face Hub API. CoolFace does not host these files.