Pixel-level occlusion detection based on sparse representation for face recognition

Zhao, Shuhuan

doi:10.1016/j.ijleo.2018.05.013

Cited by 8 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Usually, in the analysis of 2D face images, only the pixel intensity is utilised [46][47][48], whereas the spatial coordinates and relation between pixel intensity and spatial coordinates are often ignored. Herein, we propose an alternative solution for face image representation, which models analytical expressions to obtain deep insights into face patterns.…”

Section: Face Pattern Analysismentioning

confidence: 99%

Learning to represent 2D human face with mathematical model

Zhang,

Li,

Sun

et al. 2024

CAAI Trans on Intel Tech

View full text Add to dashboard Cite

How to represent a human face pattern? While it is presented in a continuous way in human visual system, computers often store and process it in a discrete manner with 2D arrays of pixels. The authors attempt to learn a continuous surface representation for face image with explicit function. First, an explicit model (EmFace) for human face representation is proposed in the form of a finite sum of mathematical terms, where each term is an analytic function element. Further, to estimate the unknown parameters of EmFace, a novel neural network, EmNet, is designed with an encoder‐decoder structure and trained from massive face images, where the encoder is defined by a deep convolutional neural network and the decoder is an explicit mathematical expression of EmFace. The authors demonstrate that our EmFace represents face image more accurate than the comparison method, with an average mean square error of 0.000888, 0.000936, 0.000953 on LFW, IARPA Janus Benchmark‐B, and IJB‐C datasets. Visualisation results show that, EmFace has a higher representation performance on faces with various expressions, postures, and other factors. Furthermore, EmFace achieves reasonable performance on several face image processing tasks, including face image restoration, denoising, and transformation.

show abstract

Section: Face Pattern Analysismentioning

confidence: 99%

Learning to represent 2D human face with mathematical model

Zhang,

Li,

Sun

et al. 2024

CAAI Trans on Intel Tech

View full text Add to dashboard Cite

show abstract

“…Occlusion can also occur intentionally; a person wearing a hat, sunglasses scarves, etc. Methods developed for dealing with occlusion include a pixel-level occlusion method [255], locally-constrained coding method [256] and a CNN based double-occlusion method in which images with occlusions are present in both training and testing [257]. The open-source face detection and alignment frameworks review revealed that typical face detection methods used are either image-based (often with neural networks) or feature-based.…”

Section: J Challengesmentioning

confidence: 99%

Open-Source Face Recognition Frameworks: A Review of the Landscape

Wanyonyi

Çelik

2022

IEEE Access

View full text Add to dashboard Cite

From holistic low-dimension feature-based segmentation to deep polynomial neural networks, Face Recognition (FR) accuracy has increased dramatically since its early days. The advancement and maturity of open-source FR frameworks have contributed to this trend, influencing many open-source research publications available in the public domain. The availability of modern accelerated computing capabilities through Graphics Process Unit (GPU) technology has played a substantial role in advancing open-source FR capabilities. The evolution and success of the open-source DL algorithms on FR, leveraging GPU technologies, have benefited from open datasets, resulting in many FR open-source implementations. This paper reviews the landscape of open-source FR frameworks, covering components of the FR pipeline across open datasets, face detection, face alignment, face representation, identification and verification, and deployment environments. We also discuss the current challenges and emerging directions in FR research.

show abstract

“…Usually, in the analysis of 2D face images, only the pixel intensity is utilized [45][46][47], whereas the spatial coordinates and relation between pixel intensity and spatial coordinates are often ignored. Herein, we propose an alternative solution for face image representation, which models analytical expressions to obtain deep insights into face patterns.…”

Section: Explicit Face Image Representation 31 Face Pattern Analysismentioning

confidence: 99%

Learning Continuous Face Representation with Explicit Functions

Zhang¹,

Li²,

Sun³

et al. 2021

Preprint

View full text Add to dashboard Cite

How to represent a face pattern? While it is presented in a continuous way in our visual system, computers often store and process the face image in a discrete manner with 2D arrays of pixels. In this study, we attempt to learn a continuous representation for face images with explicit functions. First, we propose an explicit model (EmFace) for human face representation in the form of a finite sum of mathematical terms, where each term is an analytic function element. Further, to estimate the unknown parameters of EmFace, a novel neural network, EmNet, is designed with an encoder-decoder structure and trained using the backpropagation algorithm, where the encoder is defined by a deep convolutional neural network and the decoder is an explicit mathematical expression of EmFace. Experimental results show that EmFace has a higher representation performance on faces with various expressions, postures, and other factors, compared to that of other methods. Furthermore, EmFace achieves reasonable performance on several face image processing tasks, including face image restoration, denoising, and transformation.

show abstract

Pixel-level occlusion detection based on sparse representation for face recognition

Cited by 8 publications

References 16 publications

Learning to represent 2D human face with mathematical model

Learning to represent 2D human face with mathematical model

Open-Source Face Recognition Frameworks: A Review of the Landscape

Learning Continuous Face Representation with Explicit Functions

Contact Info

Product

Resources

About