81 Hits in 7.5 sec

Gender Classification by LUT Based Boosting of Overlapping Block Patterns [chapter]

Rakesh Mehta, Manuel Günther, Sébastien Marcel
2015 Lecture Notes in Computer Science  
The paper addresses the problem of gender classification from face images.  ...  The results demonstrate that Local Binary Pattern (LBP) features with LUT boosting outperform the commonly used block-histogram-based LBP approaches and that OBP features gain over Multi-Block LBP (MB-LBP  ...  The research leading to this paper has received funding from the Swiss National Science Foundation under the National Center of Competence in Research IM2 ( and the FP7 European project BEAT  ... 
doi:10.1007/978-3-319-19665-7_45 fatcat:eryroloonzem5oqw3fotw54uwm

Real-time gender classification

Bo Wu, Haizhou Ai, Chang Huang, Hanqing Lu, Tianxu Zhang
2003 Third International Symposium on Multispectral Image Processing and Pattern Recognition  
This paper introduces an automatic real-time gender classification system. The system consists of mainly three modules, face detection, normalization and gender classification.  ...  The LUT-type weak classifier based Adaboost learning method is proposed for training both face detector and gender classifier, and a Simple Direct Appearance Model (SDAM) based method is developed to detect  ...  In addition, Fig.8 is the error rates curves of the first 250 boosting rounds of both the threshold and LUT weak classifier based Adaboost algorithm.  ... 
doi:10.1117/12.539077 fatcat:ueo3y7ecl5d27ktnztzzm6x244

LUT-Based Adaboost for Gender Classification [chapter]

Bo Wu, Haizhou Ai, Chang Huang
2003 Lecture Notes in Computer Science  
There are two main approaches to the problem of gender classification, Support Vector Machines (SVMs) and Adaboost learning methods, of which SVMs are better in correct rate but are more computation intensive  ...  This paper presents a novel Look Up Table ( LUT) weak classifier based Adaboost approach to learn gender classifier. This algorithm converges quickly and results in efficient classifiers.  ...  of both two classes largely overlap and some small peaks appear, which make it very difficult to separate them by only one threshold.  ... 
doi:10.1007/3-540-44887-x_13 fatcat:v7omrhhu6rhenmdxagqdvxme5i

A unified learning framework for object detection and classification using nested cascades of boosted classifiers

Rodrigo Verschae, Javier Ruiz-del-Solar, Mauricio Correa
2007 Machine Vision and Applications  
The proposed framework allows us to build state of the art face detection, eyes detection, and gender classification systems.  ...  In this paper a unified learning framework for object detection and classification using nested cascades of boosted classifiers is proposed.  ...  Acknowledgments This research was funded by Millenium Nucleus Center for Web Research, Grant P04-067-F, Chile.  ... 
doi:10.1007/s00138-007-0084-0 fatcat:35hcl77bfbfilc5kbofigpbk4q

Real-time embedded age and gender classification in unconstrained video

Ramin Azarmehr, Robert Laganiere, Won-Sook Lee, Christina Xu, Daniel Laroche
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
In this thesis, we present a complete framework for video-based gender classification and age estimation which can perform accurately on embedded systems in real-time and under unconstrained conditions  ...  A non-linear Support Vector Machine (SVM) classifier along with a discriminative demography-based classification strategy is exploited to improve both accuracy and performance of classification.  ...  Figure 4 .0. 1 : 41 The block diagram showing the architecture of our video-based age and gender classification system.  ... 
doi:10.1109/cvprw.2015.7301367 dblp:conf/cvpr/AzarmehrLLXL15 fatcat:5bxiwufnr5e4hfwssmomuqxr5a

Person-independent facial expression analysis by fusing multiscale cell features

Lubing Zhou, Han Wang
2013 Optical Engineering: The Journal of SPIE  
To boost noise resistance, MC-LIIP carries out comparison computation on the average values of scalable cells instead of individual pixels.  ...  A novel appearance-based feature, the multiscale cell local intensity increasing patterns (MC-LIIP), to represent facial images and conduct person-independent facial expression analysis is presented.  ...  The recognition performance can be boosted by adjusting the block partition. Table 1 lists the recognition rates of several partitions for MC-LIIP 1 image.  ... 
doi:10.1117/1.oe.52.3.037201 fatcat:xntv7rabnrg3biyllzm6tfglw4

FPGA-Based Portable Ultrasound Scanning System with Automatic Kidney Detection

R. Bharath, Punit Kumar, Chandrashekar Dusa, Vivek Akkala, Suresh Puli, Harsha Ponduri, K. Krishna, P. Rajalakshmi, S. Merchant, Mohammed Mateen, U. Desai
2015 Journal of Imaging  
Multiple windows with overlapping regions of interest are merged into a single window by averaging the coordinates of the window.  ...  Rfp(α) Angle of steering (α) Figure 13 . Calculations of delay patterns for each transducer element.  ...  Conflicts of Interest The authors declare no conflict of interest.  ... 
doi:10.3390/jimaging1010193 fatcat:rorj76gf2ja4bbsvnlxfcbyqm4

A novel SoC architecture on FPGA for ultra fast face detection

Chun He, Alexandros Papakonstantinou, Deming Chen
2009 2009 IEEE International Conference on Computer Design  
The face detector architecture extracts the coarse grained parallelism by efficiently overlapping different computation phases while taking advantage of the finegrained parallelism at the module level.  ...  One of the biggest challenges in face detection based applications is the speed at which faces can be accurately detected.  ...  FPGA-based implementation block diagram. parallelism both at the task level and the operation level.  ... 
doi:10.1109/iccd.2009.5413122 dblp:conf/iccd/HePC09 fatcat:n2ayz3opo5drpmk2bauz7fef2a

A Comprehensive Survey on Human Skin Detection

Mohammad Reza Mahmoodi, Sayed Masoud Sayedi
2016 International Journal of Image Graphics and Signal Processing  
Skin segmentation algorithms are mainly based on color information; an in-depth discussion on effectiveness of disparate color spaces is elucidated.  ...  Variety of applications in which skin detection has been either fully or partially used is also provided.  ...  SVM models Support vector machine (SVMs) are supervised learning models applied to many pattern recognition tasks as well as human skin classification.  ... 
doi:10.5815/ijigsp.2016.05.01 fatcat:rf54lquhfzdvpotdbxbbf6n5gq

A survey on face detection in the wild: Past, present and future

Stefanos Zafeiriou, Cha Zhang, Zhengyou Zhang
2015 Computer Vision and Image Understanding  
These techniques are roughly categorized into two general schemes: rigid templates, learned mainly via boosting based methods or by the application of deep neural networks, and deformable models that describe  ...  the face by its parts.  ...  Both, boosting-based and DCNNs-based algorithms take deformations and facial pose implicitly into account by selecting (boosting) or learning (DCNN) a number of features invariant to deformation and/or  ... 
doi:10.1016/j.cviu.2015.03.015 fatcat:d7ehtad5dnf5td5kvlbunp5swe

Performance Evaluation of Different Data Mining Classification Algorithm and Predictive Analysis

Syeda Farha Shazmeen Syeda Farha Shazmeen
2013 IOSR Journal of Computer Engineering  
Based on the combination of the four features, Fisher developed a linear discriminant model to distinguish the species from each other classification method to identify the class of Iris flower as Irissetosa  ...  It is used to identify hidden patterns in a large data set. Classification techniques are supervised learning techniques that classify data item into predefined class label.  ...  research because of the recent increase in the number of users in SNSs.  ... 
doi:10.9790/0661-1060106 fatcat:ffw6zxhlajaljmhstlrz6tbqhe

Automatic Face Understanding: Recognizing Families in Photos [article]

Joseph P Robinson
2021 arXiv   pre-print
Additionally, we tackled the classic problem of facial landmark localization. A majority of these networks have objectives based on L1 or L2 norms, which inherit several disadvantages.  ...  We have several benchmarks in kinship verification, family classification, tri-subject verification, and large-scale search and retrieval.  ...  We again follow the settings of [125] : resize images to 64×64, divide into 16×16 non-overlapping blocks, and use a radius of 2 and sampling number of 8.  ... 
arXiv:2102.08941v1 fatcat:eqje3jh23nb6do7crz7rw6342a

Multivariate Boosting with Look-Up Tables for Face Processing

Cosmin Atanasoaei
The unified boosting framework covers multivariate classification and regression problems and it is achieved by interpreting boosting as optimization in the functional space of the weak learners.  ...  This is achieved using look-up-tables as weak learners and multi-block Local Binary Patterns as features.  ...  Finally, we describe the binary patterns used for boosting LUTs. Thus, we introduce an efficient boosting framework to be used for a wide range of multivariate classification and regression problems.  ... 
doi:10.5075/epfl-thesis-5374 fatcat:bsl2lttdrbbmlmp2mjlfiopnoy


2020 2020 IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan)  
In particular, an entropy function of a pattern lock can be calculated, which is decided by five kinds of attributes: size, length, angle, overlap and intersection for quantitative evaluation of pattern  ...  We apply multi-output neural network paradigm to predict age and gender simultaneously and the final loss function is based on the combination of age and gender losses.  ...  Five types of patterns and eight interpolation methods of blocks are selected by the machine learning to improve the quality of demosaicking images.  ... 
doi:10.1109/icce-taiwan49838.2020.9258230 fatcat:g25vw7mzvradxna2grlzp6kgiq

A Construction Kit for Efficient Low Power Neural Network Accelerator Designs [article]

Petar Jokic, Erfan Azarkhish, Andrea Bonetti, Marc Pons, Stephane Emery, Luca Benini
2021 arXiv   pre-print
Driven by the rapid evolution of network architectures and their algorithmic features, accelerator designs are constantly updated and improved.  ...  It presents the list of optimizations and their quantitative effects as a construction kit, allowing to assess the design choices for each building block separately.  ...  and 160x120 pixel gender classification tasks.  ... 
arXiv:2106.12810v1 fatcat:gx7cspazc5fdfoi64t2zjth7am
« Previous Showing results 1 — 15 out of 81 results