Context

I picked up this dataset from https://github.com/anilsathyan7/Portrait-Segmentation. Since they did not specify the license for the dataset, I will assume that the data set is the same license as the repository (MIT). This dataset is processed, while the original comes in npy format.

Content

There are folders, the one starts with x contains the image, the one start with y contain the mask of the human face. train and test is self-explainatory. The mask images have one channel only.

Our Single-person Portrait Matting Dataset is a pivotal resource for the fashion, media, and social media industries, providing finely labeled portrait images that capture a wide range of postures and hairstyles from various countries. With a focus on high-resolution images exceeding 1080 x 1080 pixels, this dataset is tailored for applications requiring detailed segmentation, including hair, ears, fingers, and other intricate portrait features.

EasyPortrait is a 26GB large-scale dataset with 20,000 RGB images and high-quality annotated masks for advanced portrait segmentation and face parsing. It supports applications such as background removal, teeth whitening, skin enhancement, red-eye removal, and eye colorization.

Peb Ib Leeg Ib Tus Neeg Portrait Matting Dataset yog qhov tseem ceeb rau kev zam, xov xwm, thiab kev tshaj xov xwm kev lag luam, muab cov ntawv sau zoo nkauj zoo nkauj uas ntes tau ntau yam ntawm postures thiab hairstyles los ntawm ntau lub teb chaws. Nrog rau kev tsom mus rau cov duab daws teeb meem siab tshaj 1080 x 1080 pixels, cov ntaub ntawv no yog tsim los rau cov ntawv thov uas xav tau cov ncauj lus kom ntxaws segmentation, suav nrog plaub hau, pob ntseg, ntiv tes, thiab lwm yam sib txawv portrait nta.

Single-person Portrait Matting Dataset (MD-Image-003)

Introduction

Our "Single-person Portrait Matting Dataset" is a pivotal resource for the fashion, media, and social media industries, providing finely labeled portrait images that capture a wide range of postures and hairstyles from various countries. With a focus on high-resolution images exceeding 1080 x 1080 pixels, this dataset is tailored for applications requiring detailed segmentation, including hair, ears, fingers, and other intricate portrait features.

If you has interested in the full version of the datasets, featuring 50k annotated images, please visit our website maadaa.ai and leave a request.

Specification

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F22149246%2F67fcd50d42041ab8fcc41486dbf664bd%2Fsingle%20person.png?generation=1724923229039129&alt=media" alt="">

About maadaa.ai

Since 2015, maadaa.ai has been dedicated to delivering specialized AI data services. Our key offerings include:

• Data Collection: Comprehensive data gathering tailored to your needs.

• Data Annotation: High-quality annotation services for precise data labeling.

• Off-the-Shelf Datasets: Ready-to-use datasets to accelerate your projects.

• Annotation Platform: Maid-X is our data annotation platform built for efficient data annotation.

We cater to various sectors, including automotive, healthcare, retail, and more, ensuring our clients receive the best data solutions for their AI initiatives.

Matting (hair&faces) - faces dataset

Accurately estimated foreground object in images. Dataset for editing applications for creating visual effects.

💴 For Commercial Usage: To discuss your requirements, learn about the price and buy the dataset, leave a request on roman@kucev.com to buy the dataset

Content

Includes 2 folders: - images - original images of faces - masks - matting masks for images

💴 Buy the Dataset: This is just an example of the data. Leave a request on roman@kucev.com to discuss your requirements, learn about the price and buy the dataset.

keywords: head segmentation dataset, face-generation, segmentation, human faces, portrait segmentation, human face extraction, image segmentation, annotation, biometric dataset, biometric data dataset, face recognition database, facial recognition, face forgery detection, face shape, ar, augmented reality, face detection dataset, facial analysis, human images dataset, hair segmentation, matting, image matting, computer vision, deep learning, potrait matting, natural image matting

The Rotatable Portrait Display market has emerged as a vital component in various industries, providing unique solutions for dynamic content presentation. These displays, designed for the ability to switch between landscape and portrait orientations, cater to diverse applications ranging from retail advertising to d

The Segmented Portrait Masked Image of Human Faces dataset is a comprehensive collection of human face images, specifically curated for tasks involving facial recognition, segmentation, and mask detection. This dataset provides high-quality, segmented images of human faces with a focus on facial features that are either masked or unmasked. The images are meticulously annotated to facilitate advanced machine learning tasks, including but not limited to, image segmentation, facial recognition under different conditions, and mask detection.

Context

Human segmentation, i.e. high resolution extraction of humans from images, is a fascinating application with many uses. However, the problem is significantly under-constrained, making it an active area of research for developing more advanced methods. This dataset, developed by AISegment aims to help by providing a solid quality dataset of images and masks.

Quoting from the dataset author's GitHub (translated via Google Translate):

This dataset is currently the largest portrait matting dataset, containing 34,427 images and corresponding matting results. The data set was marked by the high quality of Beijing Play Star Convergence Technology Co., Ltd., and the portrait soft segmentation model trained using this data set has been commercialized.
The original images in the dataset are from Flickr, Baidu, and Taobao. After face detection and area cropping, a half-length portrait of 600*800 was generated.
The clip_img directory is a half-length portrait image in the format jpg; the matting directory is the corresponding matting file (convenient to confirm the matting quality), the format is png, you should first extract the alpha map from the png image before training. For example, using opencv you can get an alpha map like this:
In_image = cv2.imread('png image file path', cv2.IMREAD_UNCHANGED) Alpha = in_image[:,:,3]

License

See the author's GitHub.

Content

This dataset comes in two parts:
1. Full images
2. The respective RGB "masks" or "cutouts" of those images

Acknowledgements

Thanks to the folks from SegmentAI for putting this dataset together.

The Portrait Photography Services market has seen significant growth and transformation over the years, adapting to the evolving demands of customers and advancements in technology. Portrait photography serves a vital role in capturing the essence of individuals, families, and professionals, offering a means to pres

📢 Disclaimer
This dataset was created by Xuebin Qin and collaborators.
I am only re-uploading it to Kaggle for ease of use by the community.

🔗 Original GitHub Repository

U2Net Test Data

It consists of: - Images for testing the default U2Net model (salient object detection) - Images for testing the U2Net-human model (human segmentation) - Images for testing the U2Net-portrait model (human portrait generation) - The expected result images for each use case using the U2Net / U2NetP models

The Portrait Recognition Solar Simulator market is rapidly evolving as a crucial part of the solar energy industry, leveraging advanced portrait recognition technologies to optimize solar panel performance and installation processes. This innovative solution facilitates the identification of ideal solar panel placem

Masks for the CelebAMask-HQ dataset.

There are 30,000 masks for each portrait in the dataset.

The dataset was marked up using BiRefNet with trained scales on the portraits.

The "iou.csv" file contains the IoU metric between the predicted masks by the BiRefNet model and masks from the dataset. If the metric value is 0.9 or less, then there is a problem in the predicted mask.

It is assumed that the masks created using BiRefNet will be more accurate. But sometimes the original images can be bad, which can cause problems.

The AI Portrait Generator market is rapidly evolving, driven by advancements in artificial intelligence and machine learning technologies. This innovative segment leverages deep learning algorithms to create stunning, lifelike portraits from simple inputs, opening new avenues for artists, businesses, and individuals

The dataset consists of RGBA PNG images that were created using LayerDiffuse.

Each image has a dimension of 869 x 1152 x 4. This dataset can solve the problem of segmenting people and their hair.

Before generating it, I collected promts using Qwen and the "CelebAMask-HQ" dataset.

Sawiradu waa internetka Xallintu waxay u dhaxaysaa 1280 x 720 ilaa 2048 x 1080.

Avy amin'ny aterineto ny sary. Ny fanapahan-kevitra dia manomboka amin'ny 1280 x 720 ka hatramin'ny 2048 x 1080.

Discover the booming stock photo market! Explore its $5 billion valuation, 8% CAGR projection, key trends (mobile photography, diverse imagery), and leading players like Getty Images & Shutterstock. This in-depth analysis reveals growth drivers, restraints, and regional market shares, offering insights for businesses and investors.

Litšoantšo li tsoa inthaneteng. Qeto e fapana ho tloha ho 1280 x 720 ho isa ho 2048 x 1080.