Boosting Sex Identification Performance

By Shumeet Baluja, Henry A. Rowley. International Journal of Computer Vision, 2005

387 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/cc50cb74ce48c15128a0f522de415861675185b3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Age and Gender Estimation of Unfiltered Faces (2014, methodology): “Face image intensities were directly classified using SVM in [43] and later again using AdaBoost in [44].”
Understanding and Comparing Deep Neural Networks for Age and Gender Classification (2017, background): “Automated facial recognition and estimation of gender and age using machine learning models has held a high level of attention for more than two decades [21, 30, 6, 16, 13] and has become ever more relevant due to the abundance of face images on the web, and especially on social media platforms.”
TextonBoost for Image Understanding: Multi-Class Object Recognition and Segmentation by Jointly Modeling Texture, Layout, and Context (2007, methodology): “For efficiency therefore, our algorithm examines only a randomly chosen fraction ζ 1 of the possible features (Baluja and Rowley 2005).”
FaceTracer: A Search Engine for Large Collections of Images with Faces (2008, methodology): “We use the formulation of Adaboost described in [8], with the modification that errors are computed in a continuous manner (using the confidence values obtained from the SVM classifier), rather than discretely as is done in [8].”
The power of comparative reasoning (2011, background): “Several papers talk about using pairwise pixel comparisons or related quantities for pattern matching [7, 13, 14, 15, 16, 17].”
Demographic Estimation from Face Images: Human vs. Machine Performance (2015, background): “Since then, a significant progress has been made on automatic demographic estimation, and a number of approaches have been reported in the literature [8], [17], [19], [21], [23], [27], [29], [33], [39], [44], [51], [54].”
Understanding images of groups of people (2009, background): “This is lower than others [3], but our collection contains a substantial number of children, small faces and difficult expressions.”
Learning local binary patterns for gender classification on real-world face images (2012, background): “Baluja and Rowley (2007) introduced an efﬁcient gender recognition system by boosting pixel comparisons in face images.”
Evaluation of Gender Classification Methods with Automatically Detected and Aligned Faces (2008, background): “The final issue studied was the sensitivity of the gender classifiers to various detection and alignment inaccuracies as did Baluja and Rowley [14].”
More generality in efficient multiple kernel learning (2009, methodology): “Of course, other forms of boosting do not have this limitation and the state-of-the-art boosting method of (Baluja & Rowley, 2007) performs better but is still signiﬁcantly inferior to GMKL.”
Describable Visual Attributes for Face Verification and Image Search (2011, neutral): “Rowley [1] used Adaboost [21] to select a linear combination”
TextonBoost: Joint Appearance, Shape and Context Modeling for Multi-class Object Recognition and Segmentation (2006, methodology): “However, our algorithm examines only a fraction τ ¿ 1 of features, randomly chosen at each round (see [23]).”
What Else Does Your Biometric Data Reveal? A Survey on Soft Biometrics (2016, background): “In addition to the aforementioned challenges relating to the perceptual idiosyncracies of humans, there are also challenges relating to the limitations of algorithmic and hardware resources.”
Minimalistic CNN-based ensemble model for gender prediction from face images (2016, methodology): “Rather than using SVM, Baluja and Rowley (2007) used AdaBoost on raw pixels and obtained 96.40% on FERET without mixing people in training and test sets. Li et al. (2012) combined facial information with clothing and hair components obtaining 95.10% accuracy on the FERET dataset. Ullah et al. (2012)”
AFIF4: Deep Gender Classification based on AdaBoost-based Fusion of Isolated Facial Features and Foggy Faces (2017, background): “In literature, many gender classification methods are applied only to frontal or near-frontal face images [37, 38, 7, 39, 40].”
Training Hierarchical Feed-Forward Visual Recognition Models Using Transfer Learning from Pseudo-Tasks (2008, methodology): “To make a direct comparison possible, we followed the experimental setting of [ 3 ] as much as possible, and conducted experiments on the FERET data for gender recognition, where no individual is allowed to appear in the training and test simultaneously.”
Age and gender classification using convolutional neural networks (2015, methodology): “Rather than using SVM, [2] used AdaBoost for the same purpose, here again, applied to image intensities.”
Local Deep Neural Networks for gender recognition (2016, methodology): “The handcrafted features found in the literature for gender recognition can be as simple as the raw pixels (Moghaddam and Yang (2002)) or pixel differences (Baluja and Rowley (2007)).”