datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
stanford_cars
Stanford Cars Dataset
Dataset Overview
Splits:
Training: 8144 images used for model training.
Test: 8041 images used for evaluation.
Contrast: 8041 images with high contrast for robustness testing.
Gaussian Noise: 8041 images corrupted by Gaussian noise for robustness testing.
Impulse Noise: 8041 images corrupted by impulse noise for robustness testing.
JPEG Compression: 8041 compressed images for robustness testing.
Motion Blur: 8041 images with motion blur for… See the full description on the dataset page: https://huggingface.co/datasets/tanganke/stanford_cars.car-dataset-repoCarlaOcc
Database_structure
CarlaOcc/
├── CarlaOccV1/
│ ├── calib/
│ │ └── calib.yaml
│ ├── splits/
│ │ ├── test.txt
│ │ ├── train.txt
│ │ └── val.txt
│ ├── SceneMeshes/
│ │ ├── fg_actors/
│ │ ├── fg_actor_occ/
│ │ └── TownXX_Opt/
│ │ ├── bg_actors/
│ │ └── bg_actor_occ/
│ ├── TownXX_Opt_SeqXX/
│ │ ├── poses/
│ │ │ ├── cam_00.txt
│ │ │ └── lidar.txt
│ │ ├── rgb/
│ │ │ ├── image_00/
│ │ │ │ ├── 0000.png… See the full description on the dataset page: https://huggingface.co/datasets/fengyi233/CarlaOcc.IPL-CARLA-dataset
IPL-CARLA-dataset
Autonomous driving semantic segmentation dataset created with CARLA (Cars Learning to Act) simulator.
Dataset information
Images are generated from two different simulated cities. They include different weather (sunny, foggy and rainy) and daytime (morning, day, sunset and night) conditions. It contains 20000 RGB-rendered images and their corresponding ground truth segmented masks. Segmentation ground truth masks have 35 different classes with colors… See the full description on the dataset page: https://huggingface.co/datasets/isp-uv-es/IPL-CARLA-dataset.SukaSuka-image-dataset
该数据集包含了《末日时在做什么？有没有空？可以来拯救吗？》大部分主要角色角色的图像数据,来源为动漫截图与同人二创。
为方便LoRA模型训练，所有图片尺寸均截为512x640尺寸，相应打标主要由Waifu Diffusion 1.4 Tagger V2自动完成，部分手工调整。
欢迎提交PR补充或修正本数据集！
Alpha：8.21号之后的clone都是放大了两倍的图片，这是为了sdxl做准备，如果你还需要512*640尺寸的数据集，你可以在clone之后，执行下面的命令
git checkout 183e253c4c304fc6c5ef5046f1940712c349c94e
相关数据集的更正作业正在火热的进行中，请期待继续的更新吧~
Sastra_ID_Cardwds_carsCarDD
🚘 CarDD Dataset
CarDD is a novel, public, large-scale dataset specifically designed for vision-based car damage detection and segmentation.
The dataset contains 4,000 high-resolution car damage images with over 9,000 well-annotated instances, making it the largest public dataset of its kind.
The high resolution of the images (average 684,231 pixels) is a key advantage over existing datasets that have a much lower average resolution (50,334 pixels). Higher resolution allows for… See the full description on the dataset page: https://huggingface.co/datasets/harpreetsahota/CarDD.cardcaptorsakura1998
Bangumi Image Base of Card Captor Sakura (1998)
This is the image base of bangumi Card Captor Sakura (1998), we detected 59 characters, 8455 images in total. The full dataset is here.
Please note that these image bases are not guaranteed to be 100% cleaned, they may be noisy actual. If you intend to manually train models using this dataset, we recommend performing necessary preprocessing on the downloaded dataset to eliminate potential noisy samples (approximately 1%… See the full description on the dataset page: https://huggingface.co/datasets/BangumiBase/cardcaptorsakura1998.car-dataset-repo-v4cart_dataset_part1car-parts-and-damage-dataset
Car Parts and Damages Polygon Dataset
Dataset Summary
The Car Parts and Damages Polygon Dataset consists of 1,812 high-resolution images, each annotated with polygon-based segmentation masks for either car parts or car damages. The dataset is designed to support training and evaluation of deep learning models for fine-grained object detection, instance segmentation, and automotive inspection tasks.
✅ Key Stats:
Total images: 1,812
Car parts: 998 images
Car… See the full description on the dataset page: https://huggingface.co/datasets/DrBimmer/car-parts-and-damage-dataset.MathVerse-lmmseval
Dataset Card for MathVerse
This is the version for lmms-eval. This shares the same data with the official dataset.
Dataset Description
Paper Information
Dataset Examples
Leaderboard
Citation
Dataset Description
The capabilities of Multi-modal Large Language Models (MLLMs) in visual math problem-solving remain insufficiently evaluated and understood. We investigate current benchmarks to incorporate excessive visual content within textual questions, which potentially… See the full description on the dataset page: https://huggingface.co/datasets/CaraJ/MathVerse-lmmseval.ZZAMTONGlgd-cards-video-day1
LGD Cards — Day-1 PoC Video (YOLO card-pip dataset)
Auto-labeled playing-card corner-pip detection tiles from the first day of our
proof-of-concept table recordings, for Live Game Defender (LGD) — an on-prem AI integrity
monitor for live casino table games. This is the day-1 training set behind the
lgd-cards-gen2 detector.
Format: YOLO — images/{train,val} + labels/{train,val}, data.yaml (52 classes, rank+suit).
~4,630 tiles, serve-matching 2×2 tiling of the source frames.… See the full description on the dataset page: https://huggingface.co/datasets/sroot/lgd-cards-video-day1.IRIS
IRIS Dataset: Industrial Real-Sim Imagery Set
Overview
The IRIS Dataset is a comprehensive real-world dataset designed to study sim-to-real transfer for object detection in industrial robotic environments. This repository provides:
The complete real IRIS dataset: 508 annotated images of 32 mechanical components captured across four distinct, challenging industrial scenes.
Assets for synthetic data generation: All necessary 3D models, backgrounds, and materials to… See the full description on the dataset page: https://huggingface.co/datasets/Carraskito/IRIS.cardinal-benchmarkCARD-Germany-Batch1
CARD – Germany 2 Days
A comprehensive multi-modal driving dataset with stereo cameras, LiDAR, and depth annotations.
Dataset Structure
This dataset contains 28 sequences across 1 region(s):
germany_2days: 28 sequences
Data Format
Each sequence contains:
img/: Stereo camera images (cam_0, cam_1)
raw/: Raw sensor data
labels/: YOLO-format annotations
export/: Trajectory and calibration data
agg_depth/: Aggregated depth point clouds… See the full description on the dataset page: https://huggingface.co/datasets/CARD-Data/CARD-Germany-Batch1.industrial_cart_2stanford_carscar-parts-and-damage-dataset
Car Parts and Damages Polygon Dataset
Dataset Summary
The Car Parts and Damages Polygon Dataset consists of 1,812 high-resolution images, each annotated with polygon-based segmentation masks for either car parts or car damages. The dataset is designed to support training and evaluation of deep learning models for fine-grained object detection, instance segmentation, and automotive inspection tasks.
✅ Key Stats:
Total images: 1,812
Car parts: 998 images
Car… See the full description on the dataset page: https://huggingface.co/datasets/Akilarasan01/car-parts-and-damage-dataset.egyptain_cars_images_datasetCARV
CARV: A Diagnostic Benchmark for Compositional Analogical Reasoning in Multimodal LLMs
Authors: Yongkang Du, Xiaohan Zou, Minhao Cheng, Lu Lin · Pennsylvania State University
Dataset Description
CARV evaluates whether multimodal LLMs can compose transformation rules from multiple image pairs via logical set operations. Given n context pairs each depicting an atomic visual change, the model must synthesize a new rule through Union (∪), Intersection (∩), or… See the full description on the dataset page: https://huggingface.co/datasets/duyongka/CARV.MAVIS-GeometryCARLA_OOD
Dataset Card for Carla_OOD Dataset
Dataset Summary
The Carla_OOD Dataset, created using the CARLA v0.9.13 simulator, mimics the sensor configuration of the KITTI dataset with a Velodyne HDL64 LiDAR and cameras aligned with KITTI's Camera0. It is designed to advance research in anomaly detection and segmentation in autonomous systems, addressing challenges in handling unexpected multi-modal inputs.
Supported Tasks
The Carla_OOD Dataset can be used to compare… See the full description on the dataset page: https://huggingface.co/datasets/Mona4399/CARLA_OOD.V_I_D_E_OScardimagessports-cards
Digital Card Magazine Dataset
This dataset contains sports card images and their associated metadata for training machine learning models in card recognition, text extraction, and value estimation.
Dataset Description
Dataset Summary
A comprehensive collection of sports card images and metadata, including:
Front and back card images
OCR-extracted text with confidence scores
AI-analyzed card attributes
Card details (player, team, year, etc.)
Vision API labels… See the full description on the dataset page: https://huggingface.co/datasets/GotThatData/sports-cards.car-damage-dataset
Car Damage Images
A raw image collection for vehicle damage assessment. Unlabeled: these images
have no annotations yet and are intended as source material for labeling or
pre-training.
Structure
images/
001/ images_001.jpg images_002.jpg ...
002/ ...
... 683 folders
thumbnails/
001/ thumbnail_001.jpg thumbnail_002.jpg ...
... 683 folders
Branch
Files
Folders
Size… See the full description on the dataset page: https://huggingface.co/datasets/Naiscorp/car-damage-dataset.Car_Race_AI_V0 #And the Devil said, "Let there be Air"
Open-Source and free to use dataset. Only ask is to cite if you are using code or dataset for research, publication etc.
This contains 30,000,000 training timesteps using Tensorflow 2.xx and PPO algorithm to train a single agent car racing utilizing CarRacing-v3 Gymnasium environment which is a maintained fork of OpenAI’s Gym library.
This agent behaviour is non-optimal as it… See the full description on the dataset page: https://huggingface.co/datasets/privateboss/Car_Race_AI_V0.