136 datasets found

Deepfakes have become a critical social problem, and detecting them is of utmost importance. The FaceForensics++ dataset offers fake video datasets. Most of the detection...

The dataset used in the paper is Extended Yale-B, which consists of face images from 38 individuals under 9 poses and 64 illumination conditions.

The WebFace12M dataset contains web-scraped, in-the-wild images in RGB format.

The Glint360K dataset contains web-scraped, in-the-wild images in RGB format.

The WebFace4M dataset contains web-scraped, in-the-wild images in RGB format.

The IJB-B and IJB-C datasets are recognized for increased difficulty compared to the standard benchmarks.

The MORPH dataset consists of 35,276 images of 8,835 Caucasian males, 10,941 images of 2,798 Caucasian females, 56,245 images of 8,839 African-American males, and 24,855 images...

State-of-the-art deep CNN face matchers are typically created using extensive training sets of color face images. Our study reveals that such matchers attain virtually identical...

Model inversion attacks are a type of privacy attack that reconstructs private data used to train a machine learning model, solely by accessing the model.

The dataset used in the paper is VGGFace2, which is a dataset for face recognition tasks.

VGGFace is a large-scale face recognition dataset. It contains 3 million images of 9,000 different people.

Face recognition and person re-identification using paired image-attribute data, where the attributes (i.e., soft biometrics) are only available during the training phase.

Face recognition dataset used for evaluating the efficiency of the proposed Translation Variant Convolution (TVConv) operator.

Face recognition and person re-identification using paired image-attribute data, where the attributes (i.e., soft biometrics) are only available during the training phase.

A face detection benchmark with variations in pose, scale, aspect ratio, occlusion and illumination conditions.

Large scale datasets [18, 17, 27, 6] boosted text conditional image generation quality. However, in some domains it could be difficult to make such datasets and usually it could...