Penn-Fudan dataset for semantic segmentation. The dataset has been split into 146 training samples and 24 validation samples.

Corresponding blog post => Training UNet from Scratch using PyTorch

Original data set => https://www.cis.upenn.edu/~jshi/ped_html/

A dog segmentation dataset created manually typically involves the following steps:

Image selection: Selecting a set of images that include dogs in various poses and backgrounds.

Image labeling: Manually labeling the dogs in each image using a labeling tool, where each dog is segmented and assigned a unique label.

Image annotation: Annotating the labeled images with the corresponding segmentation masks, where the dog region is assigned a value of 1 and the background region is assigned a value of 0.

Dataset splitting: Splitting the annotated dataset into training, validation, and test sets.

Dataset format: Saving the annotated dataset in a format suitable for use in machine learning frameworks such as TensorFlow or PyTorch.

Dataset characteristics: The dataset may have varying image sizes and resolutions, different dog breeds, backgrounds, lighting conditions, and other variations that are typical of natural images.

Dataset size: The size of the dataset can vary, but it should be large enough to provide a sufficient amount of training data for deep learning models.

Dataset availability: The dataset may be made publicly available for research and educational purposes.

Overall, a manually created dog segmentation dataset provides a high-quality training data for deep learning models and is essential for developing robust segmentation models.

This dataset contains the necessary code for using our spray segmentation model used in the paper, ML-based semantic segmentation for quantitative spray atomization description. See README for more information.

The performance of different semantic segmentation models on the self-constructed training dataset.

Training dataset for urban green land cover and land use detection for Sentinel-2 satellite images. Samples are pixel-wise labelled scenes over the city of Prague, including bigger parks and smaller vegetation patches within high-density urban areas.

Contains four classes:

* 0: Non-vegetated pixels
* 1: Low recreational vegetation
* 2: High recreational vegetation
* 3: Non-recreational vegetation

This dataset contains the necessary code for using our spray segmentation model used in the paper, Machine learning based spray process quantification. More information can be found in the README.md.

Coast Train is a library of images of coastal environments, annotations, and corresponding thematic label masks (or ‘label images’) collated for the purposes of training and evaluating machine learning (ML), deep learning, and other models for image segmentation. It includes image sets from both geospatial satellite, aerial, and UAV imagery and orthomosaics, as well as non-geospatial oblique and nadir imagery. Images include a diverse range of coastal environments from the U.S. Pacific, Gulf of Mexico, Atlantic, and Great Lakes coastlines, consisting of time-series of high-resolution (≤1m) orthomosaics and satellite image tiles (10–30m). Each image, image annotation, and labelled image is available as a single NPZ zipped file. NPZ files follow the following naming convention: {datasource}_{numberofclasses}_{threedigitdatasetversion}.zip, where {datasource} is the source of the original images (for example, NAIP, Landsat 8, Sentinel 2), {numberofclasses} is the number of classes us ...

This dataset contains the necessary code for using our soot (instance) segmentation model used for segmenting soot filaments from PIV (Mie scattering) images. In the corresponding paper, an ablation study is conducted to delineate the effects of domain randomisation parameters of synthetically generated training data on the segmentation accuracy. The best model is used to extract high-level statistics from soot filaments in an RQL-type model combustor to enhance the fundamental understanding soot formation, transport and oxidation. B. Jose, K. P. Geigle, F. Hampp, Domain-Randomised Instance-Segmentation Benchmark for Soot in PIV Images, submitted to Machine Learning: Science and Technology (2025)

1. Convolutional neural networks (CNNs) and deep learning are powerful and robust tools for ecological applications, and are particularly suited for image data. Image segmentation (the classification of all pixels in images) is one such application and can for example be used to assess forest structural metrics. While CNN-based image segmentation methods for such applications have been suggested, widespread adoption in ecological research has been slow, likely due to technical difficulties in implementation of CNNs and lack of toolboxes for ecologists.
2. Here, we present R package imageseg which implements a CNN-based workflow for general-purpose image segmentation using the U-Net and U-Net++ architectures in R. The workflow covers data (pre)processing, model training, and predictions. We illustrate the utility of the package with image recognition models for two forest structural metrics: tree canopy density and understory vegetation density. We trained the models using large and dive...

A real-synthetic rock instance segmentation dataset for training and benchmarking.

This dataset contains fire images and their corresponding segmentation masks for a semantic segmentation task. The dataset can also be used for binary image classification task (to classify images into fire and not fire classes). The dataset has no split of train, validation, and test folders; so, for training purposes, it should be split into three sets necessary for Machine Learning and Deep Learning tasks, namely train, validation, and test splits. The structure of the data is as follows:

├── ROOT: └── images: ├── fire: ├── img_file; ├── img_file; ├── ...; └── img_file. ├── not fire: ├── img_file; ├── img_file; ├── ...; └── img_file. └── masks: ├── img_file; ├── img_file; ├── ...; └── img_file.

For the semantic segmentation task, images name and their corresponding labels have the same file name. And for the image classification task, the class labels can be obtained directory names (fire, not fire). Good luck!

This dataset contains the images and ground truth label masks for semantic segmentation created and described in "Hinniger, C.; Rüter, J. Synthetic Training Data for Semantic Segmentation of the Environment from UAV Perspective. Aerospace 2023, 10, 604. https://doi.org/10.3390/aerospace10070604".

This web-based validation system has been designed to perform visual validation of automated multi-class segmentation of concrete samples from scanning electron microscopy (SEM) images. The goal is to segment automatically SEM images into no-damage and damage sub-classes, where the damage sub-classes consist of paste damage, aggregate damage, and air voids. While the no-damage sub-classes are not included in the goal, they provide context for assigning damage sub-classes. The motivation behind this web validation system is to prepare a large number of pixel-level multi-class annotated microscopy images for training artificial intelligence (AI) based segmentation models (U-Net and SegNet models). While the purpose of the AI models is to predict accurately four damage labels, such as, paste damage, aggregate damage, air voids, and no-damage, our goal is to assert trust in such predictions (a) by using contextual labels and (b) by enabling visual validations of predicted damage labels.

Unsupervised domain adaptation for semantic segmentation via class-balanced self-training

A dataset of Landsat, Sentinel, and Planetscope satellite images of coastal shoreline regions, and corresponding semantic segmentations. The dataset consists of folders of images and label images. Label images are images where each pixel is given a discrete class by a human annotator, among the following classes: a) water, b) whitewater/surf, c) sediment, and d) other. These data are intended only to be used as a training and validation dataset for a machine learning based image segmentation model that is specifically designed for the task of coastal shoreline satellite image semantic segmentation.

The Roads Segmentation Dataset contains DVR-captured road scene images paired with pixel-level segmentation masks. The dataset includes diverse traffic scenarios, with detailed labels for roads, vehicles, signs, markings, and background objects. Each image undergoes strict quality checks to ensure accuracy, making it suitable for training computer vision models in autonomous driving, urban planning, and intelligent traffic systems.

The datasets contain RGB photos of Scots pine seedlings of three populations from two different ecotypes originating in the Czech Republic:Plasy - lowland ecotype,Trebon - lowland ecotype,Decin - upland ecotype.These photos were taken in three different periods (September 10th 2021, October 23rd 2021, January 22nd 2022).File dataset_for_YOLOv7_training.zip contains image data with annotations for training YOLOv7 segmentation model (training and validation sets)The dataset also contains a table with information on individual Scots pine seedlings:affiliation to parent tree (mum)affiliation to population (site)row and column in which the seedling was grown (row, col)affiliation to the planter in which the seedling was grown (box)mean RGB values of pine seedling in three different periods (B_september, G_september, R_september B_october, G_october, R_october, B_january, G_january, R_january)mean HSV values of pine seedling in three different periods (H_september, S_september, V_september, H_october, S_october, V_october, H_january, S_january, V_january)

F1 score of segmented nuclei images trained on an automatically and a manually annotated dataset.

The visuAAL Skin Segmentation Dataset contains 46,775 high quality images divided into a training set with 45,623 images, and a validation set with 1,152 images. Skin areas have been obtained automatically from the FashionPedia garment dataset. The process to extract the skin areas is explained in detail in the paper 'From Garment to Skin: The visuAAL Skin Segmentation Dataset'.

If you use the visuAAL Skin Segmentation Dataset, please, cite:

• https://doi.org/10.5281/zenodo.6973396
• https://doi.org/10.1007/978-3-031-13321-3_6

How to use:

1. Download the FashionPedia dataset from https://fashionpedia.github.io/home/Fashionpedia_download.html
2. Download the visuAAL Skin Segmentation Dataset. The dataset consists of two folders, namely train_masks and val_masks. Each folder corresponds to the training and validation sets in the original FashionPedia dataset.
3. After extracting the images from FashionPedia, for each image existing in the visuAAL skin segmentation dataset, the original image can be found with the same name (file_name in the annotations file).

A sample of image data in the FashionPedia dataset is:

{'id': 12305,

'width': 680,

'height': 1024,

'file_name': '064c8022b32931e787260d81ed5aafe8.jpg',

'license': 4,

'time_captured': 'March-August, 2018',

'original_url': 'https://farm2.staticflickr.com/1936/8607950470_9d9d76ced7_o.jpg',

'isstatic': 1,

'kaggle_id': '064c8022b32931e787260d81ed5aafe8'}

NOTE: Not all the images in the FashionPedia dataset have the correponding skin mask in the visuAAL Skin Segmentation Dataset, as there are images in which only garment parts and not people are present in them. These images were removed when creating the visuAAL Skin Segmentation Dataset. However, all the instances in the visuAAL skin segmentation dataset have their corresponding match in the FashionPedia dataset.

This Human Face Segmentation Dataset contains 70,846 high-quality images featuring diverse subjects with pixel-level annotations. The dataset includes individuals across various age groups—from young children to the elderly—and represents multiple ethnicities, including Asian, Black, and Caucasian. Both males and females are included. The scenes range from indoor to outdoor environments, with pure-color backgrounds also present. Facial expressions vary from neutral to complex, including large-angle head tilts, eye closures, glowers, puckers, open mouths, and more. Each image is precisely annotated on a pixel-by-pixel basis, covering facial regions, five sense organs, body parts, and appendages. This dataset is ideal for applications such as facial recognition, segmentation, and other computer vision tasks involving human face parsing.