Face Recognition by Humans: Nineteen Results All Computer Vision Researchers Should Know About

By P. Sinha, B. Balas, Yuri Ostrovsky, Richard Russell. Proceedings of the IEEE, 2006

736 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/d5c22cb54bd23f17289c31abb84baaf0cd439540

How researchers see this paper

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

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

Citations in the corpus

9 Perceiving Animacy and Intentionality: Visual Processing or Higher-Level Judgment? (2013, background): “For example, faces may automatically receive specialized processing that elbows do not, even when both are equally well attended—perhaps because the mind has specialized routines for processing faces in particular ways, whereas some other body parts may receive only domain-general visual processing ”
PsyPhy: A Psychophysics Driven Evaluation Framework for Visual Recognition (2016, background): “-style testing of models. 2D natural images are the most common type of data in computer vision, and thus form a good basis for algorithmic evaluation in this mode. For face recognition, Sinha et al. [50] suggested that the use of incremental manipulations to the resolution, pose, illumination, expr”
AdvHat: Real-World Adversarial Attack on ArcFace Face ID System (2019, background): “According to the article [29], eyebrows are the most important feature for the face recognition by a human.”
Mechanisms of face perception. (2008, neutral): “From (Sinha et al 2006).”
Face Recognition (2018, positive): “Staring at the faces in the green circles will cause one to misidentify the central face with the faces circled in red [8].”
Robust Face Recognition via Sparse Representation (2009, background): “This is partly due to the remarkable face recognition capability of the human visual system [21] and partly due to numerous important applications for face recognition technology [22].”
Co-Attentive Equivariant Neural Networks: Focusing Equivariance On Transformations Co-Occurring In Data (2019, background): “• The human visual system does not just encode familiarity to objects but seems to learn through experience the poses in which these objects customarily appear in the environment to assist and improve object recognition (Freire et al., 2000; Riesenhuber et al., 2004; Sinha et al., 2006).”
Evaluating information contributions of bottom-up and top-down processes (2009, methodology): “In our second experiment, we test human faces and cars which are two of the most studied patterns in the literature [14, 16, 13, 3, 7].”
Multiresolution Deep Belief Networks (2012, background): “The fact that high resolution is not a requirement of good recognition has been observed in human psychology [28].”
Exposing photo manipulation with inconsistent reflections (2012, background): “BACKGROUND In many ways, the human visual system is remarkable, capable of hyperacuity [Westheimer and McKee 1975], rapid scene understanding [Potter 1976], and robust face recognition [Sinha et al. 2006].”
Eye In-painting with Exemplar Generative Adversarial Networks (2017, background): “However, humans are very sensitive to small errors in facial structure, specially if those faces are our own or are well-known to us [33]; moreover, the so-called ”uncanny valley” [27] is a difficult impediment to cross when manipulating facial features.”
Deep Learning for Face Recognition: Pride or Prejudiced? (2019, background): “On the other hand, Cauca-soid subjects used traits such as the iris color, face shape, hair color and texture for describing and identifying other people of the same race [13, 33].”
Annotated Facial Landmarks in the Wild: A large-scale, real-world database for facial landmark localization (2011, background): “For humans the recognition of a familiar face is straight forward, it has even been observed that humans are able to recognize familiar faces in very low resolution images [28].”
The code for facial identity in the primate brain (2017, background): “set of transformations in articulated shape than just view changes, consistent with the invariance of face recognition behavior to many transformations beyond view changes, such as severe distortion in face aspect ratio (Sinha et al., 2006) .”
Recognizing Disguised Faces: Human and Machine Evaluation (2014, background): “Combining their results [24,30] point to a possibility that the representation of familiar faces might be more robust to certain image degradations than that of unfamiliar faces.”
Compound facial expressions of emotion (2014, background): “These fiducials define the external and internal shape of the face, because research has shown both external and internal features are important in face recognition (20).”
Perceptual Annotation: Measuring Human Vision to Improve Computer Vision (2014, background): “[3] have studied human recognition performance under challenging circumstances (low resolution, changing pose, occlusion, and various artificial distortions), emphasizing that humans perform remarkably well where W.”