FusiformNet: Extracting Discriminative Facial Features on Different Levels

11/01/2020
by   Kyo Takano, et al.
0

Over the last several years, research on facial recognition based on Deep Neural Network has evolved with approaches like task-specific loss functions, image normalization and augmentation, network architectures, etc. However, there have been few approaches with attention to how human faces differ from person to person. Premising that inter-personal differences are found both generally and locally on the human face, I propose FusiformNet, a novel framework for feature extraction that leverages the nature of person-identifying facial features. Tested on ImageUnrestricted setting of Labeled Face in the Wild benchmark, this method achieved a state-of-the-art accuracy of 96.67 normalization, or special loss functions. Likewise, the method also performed on par with previous state-of-the-arts when pretrained on CASIA-WebFace dataset. Considering its ability to extract both general and local facial features, the utility of FusiformNet may not be limited to facial recognition but also extend to other DNN-based tasks.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/31/2018

Deep Sketch-Photo Face Recognition Assisted by Facial Attributes

In this paper, we present a deep coupled framework to address the proble...
research
02/22/2023

Enhanced Face Authentication With Separate Loss Functions

The overall objective of the main project is to propose and develop a sy...
research
03/18/2019

Subjectivity and complexity of facial attractiveness

The origin and meaning of facial beauty represent a longstanding puzzle....
research
03/03/2021

DeepFN: Towards Generalizable Facial Action Unit Recognition with Deep Face Normalization

Facial action unit recognition has many applications from market researc...
research
03/20/2018

Transferring Rich Deep Features for Facial Beauty Prediction

Feature extraction plays a significant part in computer vision tasks. In...
research
08/17/2021

Neural Photofit: Gaze-based Mental Image Reconstruction

We propose a novel method that leverages human fixations to visually dec...
research
03/01/2023

Learning Person-specific Network Representation for Apparent Personality Traits Recognition

Recent studies show that apparent personality traits can be reflected fr...

Please sign up or login with your details

Forgot password? Click here to reset