datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
DailyBench
DailyBench: A Unified Benchmark for AI-Generated and Manipulated Images from Modern Generative Models
Introduction
Recent advances in generative models have shifted AI-generated image detection from identifying easily distinguishable, fully synthetic images to identifying highly realistic content generated by both modern generation and manipulation pipelines.
However, existing detection benchmarks are often built with outdated generative models and… See the full description on the dataset page: https://huggingface.co/datasets/WhiteJiangzz/DailyBench.corn-leaf-diseases-pests-and-deficiencies
Corn Leaf Diseases, Pests and Deficiencies
Dataset de imágenes de hojas de maíz para la clasificación de enfermedades, plagas y deficiencias nutricionales. Forma parte de Doctor Maíz, un sistema de diagnóstico offline pensado para pequeños agricultores de El Salvador.
Sobre este corpus se entrenó el modelo desplegado en la aplicación móvil del proyecto: EfficientNet-Lite0 cuantizado a Int8, de 3.5 MB, con macro-F1 de 0.9468 sobre un conjunto de prueba independiente de 5 015… See the full description on the dataset page: https://huggingface.co/datasets/daiv05/corn-leaf-diseases-pests-and-deficiencies.EURO_Coin_Dataset
EURO-Coin Dataset
Dataset Summary
The EURO-Coin Dataset is a collection of 12,878 images of euro coins from 23 issuing countries, spanning all 8 denominations (1 cent to 2 euro). Images were gathered via an automated web crawler that extracted publicly available pictures from Google, Bing, and Baidu image search engines.
The dataset is intended exclusively for non-commercial research purposes (e.g., computer vision, deep learning, currency recognition). The original… See the full description on the dataset page: https://huggingface.co/datasets/DAISLab-Unisa/EURO_Coin_Dataset.AIME-Dataset
AIME: AI Multimedia Ethics Dataset
The AIME (AI Multimedia Ethics) dataset is a collection of images and videos generated by
Text-to-Image (T2I) and Text-to-Video (T2V) models, manually annotated for ethical/unethical content.
Access
This dataset is gated. To access it, please request approval on this page.Access is granted for non-commercial research purposes only.
Dataset Summary
The AIME dataset was created to study the risks of generative AI… See the full description on the dataset page: https://huggingface.co/datasets/DAISLab-Unisa/AIME-Dataset.filter-inspection-data
滤光片外观质检数据集
IR-CUT 滤光片裁片的暗场成像图，来自 11 个采集包（样本7_27.zip）。
配套模型与开箱即用推理包：https://huggingface.co/daipath/filter-inspection
要上传新采集的数据？ → 上传工具/上传新数据指南.md
笔记本上跑，先体检后上传，自动剥掉 labelme 内嵌的图片副本（省约 98% 流量）。
pip install -U huggingface_hub
python upload_data.py "你的数据文件夹" # 只体检
python upload_data.py "你的数据文件夹" --yes # 真上传
📚 完整文档
文件
内容
docs/INDEX.md
项目总索引：四个仓库怎么选、三个核心结论、数据一览
docs/REPRODUCE.md
完整复现指南：环境、数据准备、逐个实验的命令与预期数字
docs/FILES.md
逐文件使用说明：41… See the full description on the dataset page: https://huggingface.co/datasets/daipath/filter-inspection-data.DaikiKase_Lora