datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
gaussian-surfels-dtuSurfaceNet-DatasetMVgaussian-surfels-bmvsKolektor_Surface_Defect
Dataset Card for Kolektor Surface-Defect Dataset
KolektorSDD (Kolektor Surface-Defect Dataset) is a grayscale industrial surface-inspection
dataset of electrical commutators.
This FiftyOne dataset uses the box-annotation release intended for the ICPR 2021 and
COMIND 2021 papers (download): one sample per
surface image, with defect regions annotated as axis-aligned bounding boxes stored as filled
rectangles in the label masks.
This is a FiftyOne dataset with 399 samples.… See the full description on the dataset page: https://huggingface.co/datasets/Voxel51/Kolektor_Surface_Defect.surfer-grpo
Kedar84/surfer-grpo
Source run file: dataset-run-1758885513916.jsonl
Generated: 2025-09-27 (UTC)
Fields:
url: Source page URL gathered by Surfer automation.
bounding_box: Normalized viewport coordinates for the target element.
raw_ss: PNG screenshot of the raw page stored in images/raw/.
annotated_ss: PNG screenshot with bounding boxes overlayed (images/annotated/).
Loading Example
from datasets import load_dataset
ds = load_dataset("Kedar84/surfer-grpo"… See the full description on the dataset page: https://huggingface.co/datasets/Kedar84/surfer-grpo.Diffuse_Map_Surfaces
Dataset Card for Diffuse Map Surface
Detailed surface textures without shadows or hotspots. (Diffuse Maps)
An LORA for Flux.1-Dev using this dataset is aviable at https://civitai.com/models/939491
dagm_defect_detection_textured_surfacesd405-hand-surface-depth-extendedwood_surface_defectsKodytek P, Bodzas A and Bilik P. A large-scale image dataset of wood surface defects for automated vision-based quality control processes [version 2; peer review: 2 approved]. F1000Research 2022, 10:581 (https://doi.org/10.12688/f1000research.52903.2)
Bounding boxes only, semantic maps
All images are 2800 x 1024 pixels (width x height)
Images compressed using
PIL.Image.Image.save(
format="JPEG",
optimize=True,
quality=50,
)
Bounding boxes converted to YOLO format.
TODO: loader… See the full description on the dataset page: https://huggingface.co/datasets/iluvvatar/wood_surface_defects.surfnerf_datad405-hand-surface-depth
D405 Hand-Surface Depth Dataset
A multi-user, multi-angle RGB-depth dataset for close-range hand-surface interaction research, captured with an Intel RealSense D405 stereo depth camera.
Paper: Real-Time Multimodal Fingertip Contact Detection via Depth and Motion Fusion for Vision-Based Human-Computer Interaction (CVPR 2026)
Authors: Mukhiddin Toshpulatov, Wookey Lee, Suan Lee, Geehyuk Lee
Key Statistics
Metric
Value
Total RGB-depth pairs
53,300… See the full description on the dataset page: https://huggingface.co/datasets/muxiddin19/d405-hand-surface-depth.weather-whiplash-surfaces
Weather Whiplash — road surface conditions
247 photographs of road and track surfaces, each labelled by a human with
one of four appearance classes. Built for Weather Whiplash, a live track-condition
detector whose reasoning layer turns a per-frame surface classification into a stable
wetness index, a trend, and a tyre-change recommendation.
The four classes
The classes describe appearance at one instant, not direction of change. There is no
"drying" class and… See the full description on the dataset page: https://huggingface.co/datasets/mrcreoid/weather-whiplash-surfaces.A-dataset-of-images-of-floating-rubbish-on-the-surface-of-the-water
License
This dataset is released under Creative Commons Attribution 4.0 International (CC BY 4.0).
Dataset: A-dataset-of-images-of-floating-rubbish-on-the-surface-of-the-water
Author: gonzz2026
URL: https://huggingface.co/datasets/gonzz2026/A-dataset-of-images-of-floating-rubbish-on-the-surface-of-the-water
License: CC BY 4.0
DOI
https://doi.org/10.57967/hf/7674
Citation
@misc{gonzz2026_floating_rubbish_2026,
title = {A Dataset of Images of… See the full description on the dataset page: https://huggingface.co/datasets/gonzz2026/A-dataset-of-images-of-floating-rubbish-on-the-surface-of-the-water.Car-Gear-Surface-Defect-Detectionmb-surface_cls
mb-surface_cls
A Mars image classification dataset for planetary science research.
Dataset Metadata
License: CC-BY-4.0 (Creative Commons Attribution 4.0 International)
Version: 1.0
Date Published: 2025-05-14
Cite As: TBD
Classes
This dataset contains the following classes:
0: apx
1: act
2: arm
3: art
4: cct
5: cio
6: clr
7: dls
8: dri
9: drh
10: drp
11: drt
12: flr
13: gro
14: hor
15: inl
16: lar
17: ltv
18: mah
19: mct
20: mas
21: mca
22: nsk
23: obt
24:… See the full description on the dataset page: https://huggingface.co/datasets/Mirali33/mb-surface_cls.synthetic-concrete-surface-normal
Synthetic Concrete Surface – Normal Images for Anomaly Detection
Dataset Description
A dataset of 747 synthetic images of defect-free concrete surfaces, generated for training one-class anomaly detection models such as PatchCore. All images are 1024×1024 PNG files depicting intact concrete with no cracks, spalling, or structural damage.
MVTec AD, the de facto benchmark for visual anomaly detection, covers 15 object and texture categories but does not include concrete.… See the full description on the dataset page: https://huggingface.co/datasets/wolfsynth/synthetic-concrete-surface-normal.mb-surface_multi_label_cls
MER - Mars Exploration Rover Dataset
A multi-label classification dataset containing Mars images from the Mars Exploration Rover (MER) mission for planetary science research.
Dataset Metadata
License: CC-BY-4.0 (Creative Commons Attribution 4.0 International)
Version: 1.0
Date Published: 2025-10-23
Cite As: TBD
Classes
This dataset uses multi-label classification, meaning each image can have multiple class labels.
The dataset contains the following classes:… See the full description on the dataset page: https://huggingface.co/datasets/Mirali33/mb-surface_multi_label_cls.sidewalk-validator-ai-dataset-surfaceproblemearth_surface_waterTestMars_surface_image_Curiosity_rover_labeled_data_set_version_1
Mars surface image (Curiosity rover) labeled data set
Authors: Alice Stanboli and Kiri L. Wagstaff
Contact: kiri.l.wagstaff@jpl.nasa.gov
This data set consists of 6691 images that were collected by the Mars
Science Laboratory (MSL, Curosity) rover by three instruments (Mastcam
Right eye, Mastcam Left eye, and MAHLI). These images are the
"browse" version of each original data product, not full resolution.
They are roughly 256x256 pixels each. Full-size images can be
obtained from… See the full description on the dataset page: https://huggingface.co/datasets/RAAArity/Mars_surface_image_Curiosity_rover_labeled_data_set_version_1.deep_pavements_surface_patchesA data-only mirror of https://github.com/kauevestena/deep_pavements_dataset
UE-Synthetic-Masonry-Surfacesmaize-leaf-health-datasetSurfaceRoughness
Surface Roughness Dataset
This is a dataset containing google image snapshots representing surface roughness categories of B, C, and D according to ASCE 7-16 26.7.2
Uses
This dataset is used for the NCSEA quick-start guide to work with machine learning models.
Data Structure
An example of data looks like
{'image': <PIL.PngImagePlugin.PngImageFile image mode=RGBA size=1350x1058>,
'label': 0}
Data Split
train
validation
test
# of… See the full description on the dataset page: https://huggingface.co/datasets/sh-zheng/SurfaceRoughness.surf-img-datasetsynthetic-masonry-surfaces
Synthetic masonry surfaces
A synthetic masonry crack dataset. This dataset is meant to be used with image segmentation models aimed at segmenting cracks in masonry surfaces, for instance using this network.
This dataset was generated using a 3D dataset generation framework for masonry surfaces and is available under the GPL-3.0 license.
Dataset structure
The dataset is split into 4058 image-label pairs. The resolution of the images is 224x224 and the labels are… See the full description on the dataset page: https://huggingface.co/datasets/DavidHidde/synthetic-masonry-surfaces.surf-img-dataset-v1wood_surface_defectsKodytek P, Bodzas A and Bilik P. A large-scale image dataset of wood surface defects for automated vision-based quality control processes [version 2; peer review: 2 approved]. F1000Research 2022, 10:581 (https://doi.org/10.12688/f1000research.52903.2)
Bounding boxes only, semantic maps
All images are 2800 x 1024 pixels (width x height)
Images compressed using
PIL.Image.Image.save(
format="JPEG",
optimize=True,
quality=50,
)
Bounding boxes converted to YOLO format.
TODO: loader… See the full description on the dataset page: https://huggingface.co/datasets/ema1995/wood_surface_defects.Surface_Quality
Dataset Card for "Surface_Quality"
More Information needed