Filters








8 Hits in 3.4 sec

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

Manuel Günther and Andras Rozsa and Terrance E. Boult
2017 arXiv   pre-print
More significantly, we introduce the Alignment-Free Facial Attribute Classification Technique (AFFACT), a data augmentation technique that allows a network to classify facial attributes without requiring  ...  Using an ensemble of three ResNets, we obtain the new state-of-the-art facial attribute classification error of 8.00% on the aligned images of the CelebA dataset.  ...  In this paper, we show that there is no need to align faces because we can use our Alignment-Free Facial Attribute Classification Technique (AFFACT) to reliably predict facial attributes based on faces  ... 
arXiv:1611.06158v2 fatcat:osdl7lqb3bg5xenyobavkj346i

Deep Facial Attribute Detection in the Wild: From General to Specific

Yuechuan Sun, Jun Yu
2018 British Machine Vision Conference  
Accurate facial attribute interpretation is a challenging task in real life due to large head poses, occlusion and illumination variations.  ...  First, we model the interdependencies of local facial regions by joint learning of all the attributes. Second, task-aware learning is established to explore the disparity regarding each attribute.  ...  AFFACT [4] proposes an alignment-free data augmentation technique using an ensemble of three ResNets for attribute classification. Kalayeh et al.  ... 
dblp:conf/bmvc/SunY18 fatcat:qzllt7f6kvb5zkntaolbmondsm

A Survey of Deep Facial Attribute Analysis [article]

Xin Zheng, Yanqing Guo, Huaibo Huang, Yi Li, Ran He
2019 arXiv   pre-print
Facial attribute analysis has received considerable attention when deep learning techniques made remarkable breakthroughs in this field over the past few years.  ...  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  ...  [31] propose an alignment-Free facial attribute classification technique (AFFACT) with data augmentation.  ... 
arXiv:1812.10265v3 fatcat:tezgo2angvfefbttuoodnss6t4

AN OVERVIEW OF FACIAL ATTRIBUTE LEARNING

Phùng Thái Thiên Trang, Fukuzawa Masayuki, Lý Quốc Ngọc
2021 Tạp chí Khoa học  
Günther et al. (2017) used ResNet to develop Alignment-Free Facial Attribute Classification Technique (AFFACT) for attribute classification on CelebA dataset.  ...  for seven facial expressions classification by using GAN.  ... 
doi:10.54607/hcmue.js.18.3.2896(2021) fatcat:trplwbymprcwbencm7err3s7vm

Aesthetics Assessment of Images Containing Faces

Simone Bianco, Luigi Celona, Raimondo Schettini
2018 2018 25th IEEE International Conference on Image Processing (ICIP)  
We exploit three different Convolutional Neural Networks to encode information regarding perceptual quality, global image aesthetics, and facial attributes; then, a model is trained to combine these features  ...  To this aim, we use the Alignment-Free Facial Attribute Classification Technique (AFFACT) [20] , shortly FA, a CNN model trained for the estimation of 40 facial attributes.  ...  We use three different CNNs to encode global image aesthetics, perceptual quality and facial attributes.  ... 
doi:10.1109/icip.2018.8451368 dblp:conf/icip/BiancoCS18 fatcat:ghr3sebfcfbazivzopol5w6uza

A Survey of Deep Facial Attribute Analysis

Xin Zheng, Yanqing Guo, Huaibo Huang, Yi Li, Ran He
2020 International Journal of Computer Vision  
Facial attribute analysis has received considerable attention when deep learning techniques made remarkable breakthroughs in this field over the past few years.  ...  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  ...  Günther et al. (2017) propose an alignment-Free facial attribute classification technique (AFFACT) with data augmentation.  ... 
doi:10.1007/s11263-020-01308-z fatcat:xmlukvd5qbenzkzjacefhcnope

A Genetic Algorithm to Combine Deep Features for the Aesthetic Assessment of Images Containing Faces

Luigi Celona, Raimondo Schettini
2021 Sensors  
Three different convolutional neural networks are exploited to encode information regarding perceptual quality, global image aesthetics, and facial attributes; then, a model is trained to combine these  ...  (b) ResNet-50 architecture used in Alignment-Free Facial Attribute Classification Technique (AFFACT) [29] . In photos containing faces, observers mainly focus on face regions.  ...  To this aim, we use the Alignment-Free Facial Attribute Classification Technique (AFFACT) [29] (FA in short), a CNN model (see the architecture in Figure 3b ) trained for the estimation of 40 facial  ... 
doi:10.3390/s21041307 pmid:33673052 pmcid:PMC7918760 fatcat:bs3zkmpqqfal3oyunwz46jy37a

Discovering multi-purpose modules through deep multitask learning [article]

Elliot Keeler Meyerson, 0000-0002-1871-2757, Austin, The University Of Texas At, Austin, The University Of Texas At, Risto Miikkulainen
2019
Machine learning scientists aim to discover techniques that can be applied across diverse sets of problems. Such techniques need to exploit regularities that are shared across tasks.  ...  Each image has binary labels for 40 facial attributes; each attribute induces a binary classification task.  ...  ; and (4) multitask visual attribute classification.  ... 
doi:10.26153/tsw/1073 fatcat:kgugc3xlkragdj4f63dc4ckhli