Filters








119,575 Hits in 8.5 sec

Multi-Domain Multi-Definition Landmark Localization for Small Datasets [article]

David Ferman, Gaurav Bharaj
2022 arXiv   pre-print
Further, we contribute a small dataset (150 images) of pareidolias to show efficacy of our method. Finally, we provide several analysis and ablation studies to justify our claims.  ...  We also show state-of-the-art performance on several varied image domain small datasets for animals, caricatures, and facial portrait paintings.  ...  We seek for our model to learn general definitions via semantic group embeddings, as an implicit facial prior, for predicting semantic group information from image feature contexts.  ... 
arXiv:2203.10358v1 fatcat:2gewm6ue2ndlnedoxokytp7rhq

Robust Head Detection in Complex Videos Using Two-stage Deep Convolution Framework

Sultan Daud Khan, Yasir Ali, Basim Zafar, Abdulfattah Noorwali
2020 IEEE Access  
Additionaly, we demonstrate that our framework achieved state-of-the-art results on four challenging benchmark datasets, i.e. HollywoodHeads, Casablanca, SHOCK, and WIDERFACE.  ...  However, the problem of scale invariance is still an open issue.  ...  The dataset has unique property of arranging the faces into three groups, i.e., small, medium and large based on face size.  ... 
doi:10.1109/access.2020.2995764 fatcat:pyifzfobbvfgffhprrtbbfrn7e

The Beauty of Capturing Faces: Rating the Quality of Digital Portraits [article]

Miriam Redi, Nikhil Rasiwasia, Gaurav Aggarwal, Alejandro Jaimes
2015 arXiv   pre-print
To this end, we procure a large dataset of face images annotated not only with aesthetic scores but also with information about the traits of the subject portrayed.  ...  We also show that a classifier trained with our features to separate beautiful portraits from non-beautiful portraits outperforms generic aesthetic classifiers.  ...  [7] : they design face-specific features and computes their effectiveness on a publicly available small-scale dataset of 150 pictures.  ... 
arXiv:1501.07304v1 fatcat:py5oxoysy5abrhquidolhz3ibi

Face Detection through Scale-Friendly Deep Convolutional Networks [article]

Shuo Yang, Yuanjun Xiong, Chen Change Loy, Xiaoou Tang
2017 arXiv   pre-print
Specifically, our method achieves 76.4 average precision on the challenging WIDER FACE dataset and 96% recall rate on the FDDB dataset with 7 frames per second (fps) for 900 * 1300 input image.  ...  We show that faces with different scales can be modeled through a specialized set of deep convolutional networks with different structures.  ...  The four splits network fails to detect small faces because of insufficient discriminative power of Res2 features to distinguish small faces from clutter background.  ... 
arXiv:1706.02863v1 fatcat:p6f55ixcpjamxibpssramgtuce

Generating images of partial face using landmark based k-nearest neighbor

Israa Hadi, Alyaa Mahdi
2020 Indonesian Journal of Electrical Engineering and Computer Science  
<p>One of the most common approaches to address the partial face recognition challenge is to crop the full face image into segments.  ...  The problem is how the full face image must be cropped in a uniform way to generate informative segments.  ...  Finally, the group of each selected key point from both of 15 and 68 detected landmark points with its k-nearest landmark points from the remaining detected landmark points is cropped to generate face  ... 
doi:10.11591/ijeecs.v17.i1.pp420-428 fatcat:m23drkkkdjafbf7dtcrzn4twsi

Image Pattern Similarity Index and Its Application to Task-Specific Transfer Learning

Jun WANG, Guoqing WANG, Leida LI
2017 IEICE transactions on information and systems  
Guided by this theory, task-specific biometric recognition model transferred from state-of-the-art DNN models is realized for both face and vein recognition.  ...  A quantized index for evaluating the pattern similarity of two different datasets is designed by calculating the number of correlated dictionary atoms.  ...  Acknowledgments This work was supported by the National Natural Science Foundation of China under Grant 61379143.  ... 
doi:10.1587/transinf.2017edl8008 fatcat:rxs2cdb4i5blva6d23rgdwu62y

EXTRACTION OF EMOTIONAL FEATURES OF HUMAN FACE

Yu He, Hongli Zhu
2021 International Journal of Applied Science and Engineering Review  
This paper introduces the face recognition dataset and the feature extraction techniques. and the deep learning method for face recognition will be introduced.  ...  by Google is used to classify face emotions, The system can recognize the face by getting the picture from the camera in real time or by reading the input face, and the accuracy rate is about 65%.  ...  The dataset contains 593 video sequences of 327 tagged expression picture sequences from 123 subjects, with subjects from an ethnically diverse group of adults.  ... 
doi:10.52267/ijaser.2021.2403 fatcat:6tyd5uzuczf4jgepxr3tkuvk2e

FAFD: Fast and Accurate Face Detector

Namho Kim, Jun-Hwa Kim, Chee Sun Won
2022 Electronics  
Specifically, based on the YOLOv5 model, we add one prediction head to increase the detection performance, especially for small faces.  ...  However, since most models focus only on the improvement of detection accuracy with computationally expensive structures, it is still far from real-time applications with a fast face detector.  ...  To this end, borrowing the size criterion from [7] , we divide the faces in the validation dataset of the WiderFace into three size groups: Small (between 10-50 pixels of the height), Medium (between  ... 
doi:10.3390/electronics11060875 fatcat:yswydso2uvggxadxo3d5k56yau

A Survey of Deep Facial Attribute Analysis [article]

Xin Zheng, Yanqing Guo, Huaibo Huang, Yi Li, Ran He
2019 arXiv   pre-print
In this paper, we provide a comprehensive survey of deep facial attribute analysis from the perspectives of both estimation and manipulation.  ...  Deep learning based facial attribute analysis consists of two basic sub-issues: facial attribute estimation (FAE), which recognizes whether facial attributes are present in given images, and facial attribute  ...  Specifically, each TSNet learns features for a specific group of attributes. Meantime, SNet shares informative features with each task.  ... 
arXiv:1812.10265v3 fatcat:tezgo2angvfefbttuoodnss6t4

Automatic Age Estimation from Face Images via Deep Ranking

Huei-Fang Yang, Bo-Yao Lin, Kuang-Yu Chang, Chu-Song Chen
2015 Procedings of the British Machine Vision Conference 2015  
This paper focuses on automatic age estimation (AAE) from face images, which amounts to determining the exact age or age group of a face image according to features from faces.  ...  Our model is with the following characteristics: (1) The scattering features are invariant to translation and small deformations. ScatNet is a deep convolutional network of specific characteristics.  ...  This paper focuses on automatic age estimation (AAE) from face images, which amounts to determining the exact age or age group of a face image according to features from faces.  ... 
doi:10.5244/c.29.55 dblp:conf/bmvc/YangLCC15 fatcat:xf6nbpynuravha6rtju6ydz7iu

Exploring Unlabeled Faces for Novel Attribute Discovery [article]

Hyojin Bahng, Sunghyo Chung, Seungjoo Yoo, Jaegul Choo
2019 arXiv   pre-print
This is a bottleneck for their real-world applications; in practice, a model trained on labeled CelebA dataset does not work well for test images from a different distribution -- greatly limiting their  ...  We aim to explore the degree to which you can discover novel attributes from unlabeled faces and perform high-quality translation.  ...  As such, small values of k produce compact clusters with highly distinctive features, while large values of k produce clusters with similar yet detailed features.  ... 
arXiv:1912.03085v1 fatcat:ylqfxvbyqbfwzmqzuosvrqx5vm

A Survey of Deep Facial Attribute Analysis

Xin Zheng, Yanqing Guo, Huaibo Huang, Yi Li, Ran He
2020 International Journal of Computer Vision  
In this paper, we provide a comprehensive survey of deep facial attribute analysis from the perspectives of both estimation and manipulation.  ...  Deep learning based facial attribute analysis consists of two basic sub-issues: facial attribute estimation (FAE), which recognizes whether facial attributes are present in given images, and facial attribute  ...  Specifically, each TSNet learns features for a specific group of attributes. Meantime, SNet shares informative features with each task.  ... 
doi:10.1007/s11263-020-01308-z fatcat:xmlukvd5qbenzkzjacefhcnope

Facial Components-based Representation for Caricature Face Recognition

Qiang Ma, Qingshan Liu
2019 International Journal of Performability Engineering  
To test the performance of our proposed representation, we build a new dataset for caricature face recognition, which consists of 259 subjects, with 6490 caricatures and 8143 photos.  ...  Each generated face contains four main facial components. Photos, caricatures, and generated faces are sent to Photo-ResNet, Caricature-ResNet, and Generated-ResNet to learn specific representations.  ...  Different from [6] , the dataset we propose consists of images with more complex head poses, even profiles.  ... 
doi:10.23940/ijpe.19.03.p5.763771 fatcat:5rp5abntwvfare5lvqcjc6vl4m

Going Deeper Into Face Detection: A Survey [article]

Shervin Minaee, Ping Luo, Zhe Lin, Kevin Bowyer
2021 arXiv   pre-print
Early approaches for face detection were mainly based on classifiers built on top of hand-crafted features extracted from local image regions, such as Haar Cascades and Histogram of Oriented Gradients.  ...  With the breakthrough work in image classification using deep neural networks in 2012, there has been a huge paradigm shift in face detection.  ...  VGGFace2 Dataset VGGFace2 dataset contains 3.31 million images of 9131 subjects, with an average of 362.6 images for each subject [83] . Images are downloaded from Google Image Search and Fig. 34 .  ... 
arXiv:2103.14983v2 fatcat:3pdac7jpvzegdnz7qzqdrs3vx4

A systemic approach to automatic metadata extraction from multimedia content

Christos Varytimidis, Georgios Tsatiris, Konstantinos Rapantzikos, Stefanos Kollias
2016 2016 IEEE Symposium Series on Computational Intelligence (SSCI)  
This system is comprised of three modules: the first provides detection of faces and recognition of known persons; the second provides generic object detection, based on a deep convolutional neural network  ...  This higher level information is used in a variety of practices, such as enriching multimedia content with external links, clickable objects and useful related information in general.  ...  The notion behind LBPH lies in the need to construct a method that can operate with small training datasets (in extreme cases, with a single training image), based on local features robust to translation  ... 
doi:10.1109/ssci.2016.7849983 dblp:conf/ssci/VarytimidisTRK16 fatcat:txwake6yvnc65f26exnz2tma3q
« Previous Showing results 1 — 15 out of 119,575 results