Face perception | EPFL Graph Search

Facial perception is an individual's understanding and interpretation of the face. Here, perception implies the presence of consciousness and hence excludes automated facial recognition systems. Although facial recognition is found in other species, this article focuses on facial perception in humans. The perception of facial features is an important part of social cognition. Information gathered from the face helps people understand each other's identity, what they are thinking and feeling, anticipate their actions, recognize their emotions, build connections, and communicate through body language. Developing facial recognition is a necessary building block for complex societal constructs. Being able to perceive identity, mood, age, sex, and race lets people mold the way we interact with one another, and understand our immediate surroundings. Though facial perception is mainly considered to stem from visual intake, studies have shown that even people born blind can learn face perc

On the Recognition Performance of BioHashing on state-of-the-art Face Recognition models

Sébastien Marcel, Hatef Otroshi Shahreza

Face recognition has become a popular authentication tool in recent years. Modern state-of-the-art (SOTA) face recognition methods rely on deep neural networks, which extract discriminative features from face images. Although these methods have high recognition performance, the extracted features contain privacy-sensitive information. Hence, the users' privacy would be jeopardized if the features stored in the face recognition system were compromised. Accordingly, protecting the extracted face features (templates) is an essential task in face recognition systems. In this paper, we use BioHashing for face template protection and aim to establish the minimum BioHash length that would be required in order to maintain the recognition accuracy achieved by the corresponding unprotected system. We consider two hypotheses and experimentally show that the performance depends on the value of the BioHash length (as opposed to the ratio of the BioHash length to the dimension of the original features). To eliminate bias in our experiments, we use several SOTA face recognition models with different network structures, loss functions, and training datasets, and we evaluate these models on two different datasets (LFW and MOBIO). We provide an open-source implementation of all the experiments presented in this paper so that other researchers can verify our findings and build upon our work.

IEEE2021

The missing social instinct

Nicole Regula Zürcher Wimmer

Smooth social interactions in everyday life rely on spontaneous understanding of emotions and intentions of others, conveyed by subtle social cues, such as gaze and facial expressions. In individuals with autism, however, who represent close to 1% of the population, this effortless and spontaneous understanding is impaired, and atypical eye contact as well as difficulties in social interaction are central to the condition. A lack of social instinct has been suggested as the essence of this neurodevelopmental disorder. In this work, I investigated configural face processing, interactions of emotions and eye gaze direction as well as empathy for pain in high- functioning young adults with autism spectrum disorders (ASD) using magnetic resonance imaging and behavioral measures. I also conducted anatomical investigations of white matter in adolescents and adults with autism to investigate structural differences underlying socio-cognitive processing. Findings showed that individuals with ASD process faces in a qualitatively similar manner as controls, namely configurally, as a unified whole. Moreover, the results showed that when participants are explicitly cued to the eye region, activation of cortical areas involved in social and emotional aspects of face processing reaches levels comparable to those observed in controls. When confronted with quick social gaze cues signaling an environmental threat, however, participants with ASD failed to show activation in areas associated with social attention, including brain regions driving eye gaze and emotion processing as well as the attribution of mental states. In contrast, affective empathy, referring to emotional contagion that has us “feeling like others” is preserved in ASD when viewing facial expressions of physical pain in others. Cognitive empathy on the other hand, which is associated with the attribution of mental states to others as well as top-down regulation of emotional arousal, seems impaired in ASD. In addition, across experiments, direct eye contact led to hyperactivation of subcortical regions suggesting hyperarousal in response to direct gaze that could underlie gaze avoidance observed in autism. Finally, we observed signs of delayed maturation in white matter tracts involved in emotional face processing in adolescents with ASD, but not in adults, which could be related to the improvements often observed in social interactions as individuals with autism enter adulthood. The studies conducted here shed light onto neural processes underlying social cognition deficits in ASD and provide clues for improving behavioral therapies.

EPFL2012

Multivariate Boosting with Look-Up Tables for Face Processing

Cosmin Atanasoaei

This thesis proposes a novel unified boosting framework. We apply this framework to the several face processing tasks, face detection, facial feature localisation, and pose classification, and use the same boosting algorithm and the same pool of features (local binary features). This is in contrast with the standard approaches that make use of a variety of features and models, for example AdaBoost, cascades of boosted classifiers and Active Appearance Models. 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. Thus a wide range of smooth loss functions can be optimized with the same algorithm. There are two general optimization strategies we propose that extend recent works on TaylorBoost and Variational AdaBoost. The first proposition is an empirical expectation formulation that minimizes the average loss and the second is a variational formulation that includes an additional penalty for large variations between predictions. These two boosting formulations are used to train real-time models using local binary features. This is achieved using look-up-tables as weak learners and multi-block Local Binary Patterns as features. The resulting boosting algorithms are simple, efficient and easily scalable with the available resources. Furthermore, we introduce a novel coarse-to-fine feature selection method to handle high resolution models and a bootstrapping algorithm to sample representative training data from very large pools of data. The proposed approach is evaluated for several face processing tasks. These tasks include frontal face detection (binary classification), facial feature localization (multivariate regression) and pose estimation (multivariate classification). Several studies are performed to assess different optimization algorithms, bootstrapping parametrizations and feature sharing methods (for the multivariate case). The results show good performance for all of these tasks. In addition to this, two other contributions are presented. First, we propose a context-based model for removing the false alarms generated by a given generic face detector. Second, we propose a new face detector that predicts the Jaccard distance between the current location and the ground truth. This allows us to formulate the face detection problem as a regression task.

EPFL2012

Multivariate Boosting with Look-Up Tables for Face Processing

Cosmin Atanasoaei

EPFL2012

The missing social instinct

Nicole Regula Zürcher Wimmer

EPFL2012