scholarly journals Subject Identification Across Large Expression Variations Using 3D Facial Landmarks

2021 ◽  
pp. 5-13
Author(s):  
SK Rahatul Jannat ◽  
Diego Fabiano ◽  
Shaun Canavan ◽  
Tempestt Neal
Keyword(s):  
Facial Landmarks ◽  
Subject Identification

Exploring the Facial Color Representative Regions Using the Humanae Images

2020 ◽  
Author(s):  
Yuchun Yan ◽  
Hayan Choi ◽  
Hyeon-Jeong Suk
Keyword(s):  
Image Analysis ◽  
Future Study ◽  
Skin Color ◽  
Color Difference ◽  
Facial Skin ◽  
Female Images ◽  
Gender And Race ◽  
Facial Landmarks

It is difficult to describe facial skin color through a solid color as it varies from region to region. In this article, the authors utilized image analysis to identify the facial color representative region. A total of 1052 female images from Humanae project were selected as a solid color was generated for each image as their representative skin colors by the photographer. Using the open CV-based libraries, such as EOS of Surrey Face Models and DeepFace, 3448 facial landmarks together with gender and race information were detected. For an illustrative and intuitive analysis, they then re-defined 27 visually important sub-regions to cluster the landmarks. The 27 sub-region colors for each image were finally derived and recorded in L ∗ , a ∗ , and b ∗ . By estimating the color difference among representative color and 27 sub-regions, we discovered that sub-regions of below lips (low Labial) and central cheeks (upper Buccal) were the most representative regions across four major ethnicity groups. In future study, the methodology is expected to be applied for more image sources.

Driver’s Drowsiness Detection Using Dlib and IOT

2020 ◽  
Vol 10 ◽  
Author(s):  
Prasanna Lakshmi Kompalli ◽  
Padma Vallakati ◽  
Ganapathi Raju Nadimpalli ◽  
Vinod Mahesh Jain ◽  
Samuel Annepogu
Keyword(s):  
Aspect Ratio ◽  
The Other ◽  
Human Intervention ◽  
Road Accidents ◽  
Online Content ◽  
Facial Landmarks ◽  
Drowsiness Detection ◽  
Rear Part ◽  
Cause Of Deaths ◽  
Published Paper

Background: Road accidents are major cause of deaths worldwide. This is enormously due to fatigue, drowsiness and microsleep of the drivers. This don’t just risk the life of driver and copassengers but also a great threat to the vehicles and humans moving around that vehicle. Methods: Research, online content and previously published paper related to drowsiness are reviewed. Using the facial landmarks DAT file, the prototype will locate and get the eye coordinates and it will calculate Eye Aspect Ratio (EAR). The EAR indicates whether the driver is drowsy or not based on the result various sensors gets activated such as Alarm generator, LED Indicators, LCD message scroll, message sent to owner and engine gets locked. Results: The prototype is able to locate eyes in the frame and detect whether the person is sleepy or not. Whenever the person is feeling drowsy alarm gets generated in the cabinet on further if the person is feeling drowsy, LED indicators will start glowing, messaging will be scrolling at the rear part of vehicle so that other vehicles and humans gets cautioned and vehicle slows down and engine gets locked. Conclusion: This prototype will help in reduction of road accidents due to human intervention. It is not only helpful to the person who install it in their vehicles but also for the other vehicles and humans moving around it.

scholarly journals Guided Spatial Transformers for Facial Expression Recognition

2021 ◽  
Vol 11 (16) ◽  
pp. 7217
Author(s):  
Cristina Luna-Jiménez ◽  
Jorge Cristóbal-Martín ◽  
Ricardo Kleinlein ◽  
Manuel Gil-Martín ◽  
José M. Moya ◽  
...  
Keyword(s):  
Facial Expression ◽  
Facial Expression Recognition ◽  
Expert Knowledge ◽  
Visual Saliency ◽  
Expression Recognition ◽  
Spatial Transformations ◽  
Facial Landmarks ◽  
Saliency Maps ◽  
Best Fit

Spatial Transformer Networks are considered a powerful algorithm to learn the main areas of an image, but still, they could be more efficient by receiving images with embedded expert knowledge. This paper aims to improve the performance of conventional Spatial Transformers when applied to Facial Expression Recognition. Based on the Spatial Transformers’ capacity of spatial manipulation within networks, we propose different extensions to these models where effective attentional regions are captured employing facial landmarks or facial visual saliency maps. This specific attentional information is then hardcoded to guide the Spatial Transformers to learn the spatial transformations that best fit the proposed regions for better recognition results. For this study, we use two datasets: AffectNet and FER-2013. For AffectNet, we achieve a 0.35% point absolute improvement relative to the traditional Spatial Transformer, whereas for FER-2013, our solution gets an increase of 1.49% when models are fine-tuned with the Affectnet pre-trained weights.

The Development of an Identification Photo Booth System based on a Deep Learning Automatic Image Capturing Method

2020 ◽  
Author(s):  
Yu-Xiang Zhao ◽  
Yi-Zeng Hsieh ◽  
Shih-Syun Lin
Keyword(s):  
Deep Learning ◽  
Experimental Results ◽  
Automatic Annotation ◽  
Learning Method ◽  
Facial Region ◽  
Facial Landmarks ◽  
Image Capturing ◽  
The Face ◽  
Facial Contours

With advances in technology, photo booths equipped with automatic capturing systems have gradually replaced the identification (ID) photo service provided by photography studios, thereby enabling consumers to save a considerable amount of time and money. Common automatic capturing systems employ text and voice instructions to guide users in capturing their ID photos; however, the capturing results may not conform to ID photo specifications. To address this issue, this study proposes an ID photo capturing algorithm that can automatically detect facial contours and adjust the size of captured images. The authors adopted a deep learning method (You Only Look Once) to detect the face and applied a semi-automatic annotation technique of facial landmarks to find the lip and chin regions from the facial region. In the experiments, subjects were seated at various distances and heights for testing the performance of the proposed algorithm. The experimental results show that the proposed algorithm can effectively and accurately capture ID photos that satisfy the required specifications.

Using facial landmarks to assist the stereo matching in fringe projection based 3D face profilometry

2021 ◽  
Vol 60 (07) ◽  
Author(s):  
Yanqiong Guo ◽  
Jiangping Zhu ◽  
Hailong Jing ◽  
Pei Zhou ◽  
Ning Xiao
Keyword(s):  
Stereo Matching ◽  
Fringe Projection ◽  
3D Face ◽  
Facial Landmarks

Preliminary Evaluation of an Electromagnetic Three-Dimensional Digitizer in Facial Anthropometry

1998 ◽  
Vol 35 (1) ◽  
pp. 9-15 ◽  
Author(s):  
Virgilio F. Ferrario ◽  
Chiarella Sforza ◽  
Carlo E. Poggio ◽  
Massimiliano Cova ◽  
Gianluca Tartaglia
Keyword(s):  
Three Dimensional ◽  
Preliminary Evaluation ◽  
Men And Women ◽  
Facial Landmarks ◽  
Coefficients Of Variation ◽  
The Face ◽  
Human Adults ◽  
Facial Anthropometry ◽  
Sufficient Precision

Objective In this investigation, the precision of a commercial three-dimensional digitizer in the detection of facial landmarks in human adults was assessed. Methods Fifty landmarks were identified and marked on the faces of five men, on five women, and on a stone cast of the face of one man. For each subject, the three-dimensional coordinates of the landmarks were obtained twice using an electromagnetic three-dimensional digitizer, and the duplicate digitizations were superimposed using common orientations and centers of gravity. Metric differences between homologous landmarks were assessed, and Dahlberg's error was computed. Results For both men and women, the error was 1.05% of the nasion-mid-tragion distance, while for the cast, it was 0.9%. When the duplicate digitizations were used to mathematically reconstruct the faces, and several distances, angles, volumes, and surfaces were computed, more than 80% of the measurements had coefficients of variation lower than 1%. Conclusions The digitizer can assess the coordinates of facial landmarks with sufficient precision, and reliable measurements can be obtained.

Recognition of emotions towards commercial ads based on EEG and facial landmarks

2019 ◽  
pp. 99-106
Author(s):  
E.V. Empaynado ◽  
H.W. Francisco ◽  
N.K. Gardose ◽  
J.J. Azcarraga
Keyword(s):  
Facial Landmarks ◽  
Recognition Of Emotions

Serial Observation of Asymmetry in the Growing Face

1992 ◽  
Vol 19 (4) ◽  
pp. 273-285 ◽  
Author(s):  
P. H. Burke
Keyword(s):  
Three Dimensional ◽  
Facial Asymmetry ◽  
Positive Sign ◽  
Growth Spurt ◽  
Facial Landmarks ◽  
Osseous Dysplasia ◽  
Dimensional Measurements ◽  
Post Traumatic ◽  
The Asymmetry ◽  
Serial Observation

Three children suffering from facial asymmetry were observed annually using facial stereophotogrammetry before, during, and after their general skeletal adolescent growth spurt. Stereophotogrammetry allows accurate three-dimensional measurements between identifiable facial landmarks. Five pairs of bilateral parameters connecting external canthi and angles of the mouth to alae and tip of nose, and to each other, allowed a positive sign (right-side larger) or a negative (left-side larger) assessment of parameter asymmetry, Their total, taking sign into account, assessed mid-facial asymmetry. Serial observation showed that: (1) in patient no. 1 suffering from post-traumatic condylar hypoplasia, the facial asymmetry resolved; (2) in patient no. 2 suffering from unilateral facial hypoplasia, the asymmetry, which was severe, reduced with adolescence, but did not resolve; (3) in patient no. 3 suffering from fibro-osseous dysplasia of left maxilla, the asymmetry was reduced by surgery, but the full effects of the surgery were not measurable until over 1 year after operation: subsequently, the asymmetry began to increase again.

Sign in / Sign up

Export Citation Format

Share Document