Filters








392,955 Hits in 5.6 sec

ByeGlassesGAN: Identity Preserving Eyeglasses Removal for Face Images [article]

Yu-Hui Lee, Shang-Hong Lai
2020 arXiv   pre-print
The encoder is responsible for extracting information from the source face image, and the face decoder utilizes this information to generate glasses-removed images.  ...  Furthermore, we demonstrate significant improvement in face recognition accuracy for face images with glasses by applying our method as a pre-processing step in our face recognition experiment.  ...  CelebA is a dataset containing 202,599 images of 10,177 celebrities and annotated with 40 face attribute labels for each image.  ... 
arXiv:2008.11042v1 fatcat:c4gejbrkcncthdw4l655qkh2ae

Taking Modality-free Human Identification as Zero-shot Learning [article]

Zhizhe Liu, Xingxing Zhang, Zhenfeng Zhu, Shuai Zheng, Yao Zhao, Jian Cheng
2020 arXiv   pre-print
There have been numerous methods proposed for human identification, such as face identification, person re-identification, and gait identification.  ...  In addition, the semantics-guided spatial attention is enforced on visual modality to obtain representations with both high global category-level and local attribute-level discrimination.  ...  ., face identification and re-ID. A. Modality-Free Face Identification Datasets. Among the most widely used datasets for face identification, we select two attribute datasets.  ... 
arXiv:2010.00975v1 fatcat:p5vx4d4msjdgxm3n3d4rbcsuim

Stereotype-Free Classification of Fictitious Faces [article]

Mohammadhossein Toutiaee, Soheyla Amirian, John A. Miller, Sheng Li
2020 arXiv   pre-print
We present a novel approach through penalized regression to label stereotype-free GAN-generated synthetic unlabeled images.  ...  GAN-made faces would be exposed to such discrimination, if they are classified by human perception.  ...  In general, we expect that this study for stereotype-free labeling of the GAN-made faces will provide interesting avenues for future work.  ... 
arXiv:2005.02157v1 fatcat:olyeawmfffaozasllmegwvpefq

SemanticAdv: Generating Adversarial Examples via Attribute-conditional Image Editing [article]

Haonan Qiu, Chaowei Xiao, Lei Yang, Xinchen Yan, Honglak Lee, Bo Li
2020 arXiv   pre-print
To further demonstrate the applicability of SemanticAdv beyond face recognition domain, we also generate semantic perturbations on street-view images.  ...  In this paper, we aim to explore the impact of semantic manipulation on DNNs predictions by manipulating the semantic attributes of images and generate "unrestricted adversarial examples".  ...  In total, for each image x, we generate 17 adversarial images with different augmented attributes.  ... 
arXiv:1906.07927v4 fatcat:tyduj5qtsjhcbiyeqhybb3xhmm

Face De-occlusion using 3D Morphable Model and Generative Adversarial Network [article]

Xiaowei Yuan, In Kyu Park
2019 arXiv   pre-print
In this paper, a novel method is proposed to restore de-occluded face images based on inverse use of 3DMM and generative adversarial network.  ...  The 3DMM serves not only as geometric prior but also proposes the face region for the local discriminator.  ...  The generator takes the occluded image and 3DMM synthesis as input to generate the occlusion-free image.  ... 
arXiv:1904.06109v2 fatcat:o2rukaqayjemvhsvjqfnewxyvm

LEED: Label-Free Expression Editing via Disentanglement [article]

Rongliang Wu, Shijian Lu
2020 arXiv   pre-print
The idea is to disentangle the identity and expression of a facial image in the expression manifold, where the neutral face captures the identity attribute and the displacement between the neutral image  ...  and the expressive image captures the expression attribute.  ...  Learning Interpolator: With the identity attribute C I N of the input image and the expression attribute ∆exp R B of the reference image, we can easily obtain the latent code C I B for the target image  ... 
arXiv:2007.08971v1 fatcat:wtbaxrdenndxxgahoe7tsn75ay

Multi-Attributed and Structured Text-to-Face Synthesis [article]

Rohan Wadhawan, Tanuj Drall, Shubham Singh, Shampa Chakraverty
2021 arXiv   pre-print
Text-to-image synthesis, especially text-to-face synthesis, has promising use cases of robust face-generation from eye witness accounts and augmentation of the reading experience with visual cues.  ...  We also consolidate a Multi-Attributed and Structured Text-to-face (MAST) dataset consisting of high-quality images with structured textual annotations and make it available to researchers to experiment  ...  Fig. 6 : 6 Some Generated Face images along with their textual descriptions. Fig. 7 : 7 Some Generated Face images along with their textual descriptions.  ... 
arXiv:2108.11100v1 fatcat:mxllb23jc5a5vg6m4srww5gulm

Deep Attributes for One-Shot Face Recognition [chapter]

Aishwarya Jadhav, Vinay P. Namboodiri, K. S. Venkatesh
2016 Lecture Notes in Computer Science  
While face recognition has considered the use of attribute based representations, for one-shot face recognition, the methods proposed so far have been using different features that represent the limited  ...  We postulate that by using an intermediate attribute representation, it is possible to outperform purely face based feature representation for one-shot recognition.  ...  by VGG-face and attribute descriptors with Exemplar-SVM (E-SVM), SVM with free- scale LDA OSS kernel and exponential of OSS kernel.  ... 
doi:10.1007/978-3-319-49409-8_44 fatcat:njd7265llnas7bmuer5zd76e6m

Deep Feature Interpolation for Image Content Changes [article]

Paul Upchurch, Jacob Gardner, Geoff Pleiss, Robert Pless, Noah Snavely, Kavita Bala, Kilian Weinberger
2017 arXiv   pre-print
We propose Deep Feature Interpolation (DFI), a new data-driven baseline for automatic high-resolution image transformation.  ...  This is particularly unexpected as DFI requires no specialized network architecture or even any deep network to be trained for these tasks.  ...  We use the Labeled Faces in the Wild (LFW) data set, which contains 13,143 images of faces with predicted annotations for 73 different attributes (e.g., SUNGLASSES, GENDER, ROUND FACE, CURLY HAIR, MUSTACHE  ... 
arXiv:1611.05507v2 fatcat:ar53k6agnbbadpket6vptwqxvy

Deep Feature Interpolation for Image Content Changes

Paul Upchurch, Jacob Gardner, Geoff Pleiss, Robert Pless, Noah Snavely, Kavita Bala, Kilian Weinberger
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We propose Deep Feature Interpolation (DFI), a new datadriven baseline for automatic high-resolution image transformation.  ...  This is particularly unexpected as DFI requires no specialized network architecture or even any deep network to be trained for these tasks.  ...  We use the Labeled Faces in the Wild (LFW) data set, which contains 13,143 images of faces with predicted annotations for 73 different attributes (e.g., SUNGLASSES, GENDER, ROUND FACE, CURLY HAIR, MUSTACHE  ... 
doi:10.1109/cvpr.2017.645 dblp:conf/cvpr/UpchurchGPPSBW17 fatcat:7xqscf7ogne2nazr7dbx36f53e

AFFACT - Alignment-Free Facial Attribute Classification Technique [article]

Manuel Günther and Andras Rozsa and Terrance E. Boult
2017 arXiv   pre-print
We show that this approach outperforms the CelebA baseline on unaligned images with a relative improvement of 36.8%.  ...  alignment beyond detected face bounding boxes.  ...  Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright annotation thereon.  ... 
arXiv:1611.06158v2 fatcat:osdl7lqb3bg5xenyobavkj346i

A State-of-the-Art Review on Image Synthesis with Generative Adversarial Networks

Lei Wang, Wei Chen, Wenjia Yang, Fangming Bi, Fei Richard Yu
2020 IEEE Access  
Then, we discuss the challenges faced by GANs and introduce some methods to deal with these problems.  ...  INDEX TERMS Generative adversarial networks, image synthesis, image-to-image translation, image editing, cartoon generation.  ...  with free-form masks anywhere or any shapes.  ... 
doi:10.1109/access.2020.2982224 fatcat:p5uxjh4cybfw5grp6ldhkpukrm

MaskMTL: Attribute prediction in masked facial images with deep multitask learning [article]

Prerana Mukherjee, Vinay Kaushik, Ronak Gupta, Ritika Jha, Daneshwari Kankanwadi, Brejesh Lall
2022 arXiv   pre-print
Predicting attributes in the landmark free facial images is itself a challenging task which gets further complicated when the face gets occluded due to the usage of masks.  ...  The source code is available at https://github.com/ritikajha/Attribute-prediction-in-masked-facial-images-with-deep-multitask-learning  ...  In this paper, we propose a landmark free end-to-end deep multi-task learning approach for masked facial attribute prediction.  ... 
arXiv:2201.03002v2 fatcat:cukwzfuzojgnnee6s6chypghaa

Exploiting video sequences for unsupervised disentangling in generative adversarial networks [article]

Facundo Tuesca, Lucas C. Uzal
2019 arXiv   pre-print
For the considered datasets, this is the identity of the generated face. The other subspace controls motion attributes, those attributes that are observed to change along short videos.  ...  In this work we present an adversarial training algorithm that exploits correlations in video to learn --without supervision-- an image generator model with a disentangled latent space.  ...  There are also some attempts to exploit video datasets for disentangling motion attributes from content attributes for video generation.  ... 
arXiv:1910.11104v1 fatcat:n23j35nasbgtnjw2ki2o26qaj4

Unsupervised Eyeglasses Removal in the Wild [article]

Bingwen Hu, Zhedong Zheng, Ping Liu, Wankou Yang, Mingwu Ren
2020 arXiv   pre-print
Given two facial images with and without eyeglasses, the proposed model learns to swap the eye area in two faces.  ...  The proposed method does not depend on the dense annotation of eyeglasses location but benefits from the large-scale face images with weak annotations.  ...  In particular, G X generates face images without glasses and G Y generates face images with glasses, respectively. Discriminator. (D X , D Y ) are two discriminators for X and Y, respectively.  ... 
arXiv:1909.06989v4 fatcat:meigab4njrf2jm54sujzodazzi
« Previous Showing results 1 — 15 out of 392,955 results