Automated facial coding software outperforms people in recognizing neutral faces as neutral from standardized datasets.
Ontology highlight
ABSTRACT:
SUBMITTER: Lewinski P
PROVIDER: S-EPMC4565996 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Face verification, though an easy task for humans, is a long-standing open research area. This is largely due to the challenging covariates, such as disguise and aging, which make it very hard to accurately verify the identity of a person. This paper investigates human and machine performance for recognizing/verifying disguised faces. Performance is also evaluated under familiarity and match/mismatch with the ethnicity of observers. The findings of this study are used to develop an automated algorithm to verify the faces presented under disguise variations. We use automatically localized feature descriptors which can identify disguised face patches and account for this information to achieve improved matching accuracy. The performance of the proposed algorithm is evaluated on the IIIT-Delhi Disguise database that contains images pertaining to 75 subjects with different kinds of disguise variations. The experiments suggest that the proposed algorithm can outperform a popular commercial system and evaluates them against humans in matching disguised face images.
| S-EPMC4100743 | biostudies-literature
Project description:Electroenchephalography (EEG) recordings collected with developmental populations present particular challenges from a data processing perspective. These EEGs have a high degree of artifact contamination and often short recording lengths. As both sample sizes and EEG channel densities increase, traditional processing approaches like manual data rejection are becoming unsustainable. Moreover, such subjective approaches preclude standardized metrics of data quality, despite the heightened importance of such measures for EEGs with high rates of initial artifact contamination. There is presently a paucity of automated resources for processing these EEG data and no consistent reporting of data quality measures. To address these challenges, we propose the Harvard Automated Processing Pipeline for EEG (HAPPE) as a standardized, automated pipeline compatible with EEG recordings of variable lengths and artifact contamination levels, including high-artifact and short EEG recordings from young children or those with neurodevelopmental disorders. HAPPE processes event-related and resting-state EEG data from raw files through a series of filtering, artifact rejection, and re-referencing steps to processed EEG suitable for time-frequency-domain analyses. HAPPE also includes a post-processing report of data quality metrics to facilitate the evaluation and reporting of data quality in a standardized manner. Here, we describe each processing step in HAPPE, perform an example analysis with EEG files we have made freely available, and show that HAPPE outperforms seven alternative, widely-used processing approaches. HAPPE removes more artifact than all alternative approaches while simultaneously preserving greater or equivalent amounts of EEG signal in almost all instances. We also provide distributions of HAPPE's data quality metrics in an 867 file dataset as a reference distribution and in support of HAPPE's performance across EEG data with variable artifact contamination and recording lengths. HAPPE software is freely available under the terms of the GNU General Public License at https://github.com/lcnhappe/happe.
| S-EPMC5835235 | biostudies-literature
Project description:DamID is a powerful technique for identifying regions of the genome bound by a DNA-binding (or DNA-associated) protein. Currently no method exists for automatically processing next-generation sequencing DamID (DamID-seq) data, and the use of DamID-seq datasets with normalisation based on read-counts alone can lead to high background and the loss of bound signal. DamID-seq thus presents novel challenges in terms of normalisation and background minimisation. We describe here damidseq_pipeline, a software pipeline that performs automatic normalisation and background reduction on multiple DamID-seq FASTQ or BAM datasets. Single replicate profiling of pol II occupancy in 3rd instar larval neuroblasts of Drosophila
2015-06-26 | E-GEOD-69184 | biostudies-arrayexpress
Project description:Emotional facial expressions can inform researchers about an individual's emotional state. Recent technological advances open up new avenues to automatic Facial Expression Recognition (FER). Based on machine learning, such technology can tremendously increase the amount of processed data. FER is now easily accessible and has been validated for the classification of standardized prototypical facial expressions. However, applicability to more naturalistic facial expressions still remains uncertain. Hence, we test and compare performance of three different FER systems (Azure Face API, Microsoft; Face++, Megvii Technology; FaceReader, Noldus Information Technology) with human emotion recognition (A) for standardized posed facial expressions (from prototypical inventories) and (B) for non-standardized acted facial expressions (extracted from emotional movie scenes). For the standardized images, all three systems classify basic emotions accurately (FaceReader is most accurate) and they are mostly on par with human raters. For the non-standardized stimuli, performance drops remarkably for all three systems, but Azure still performs similarly to humans. In addition, all systems and humans alike tend to misclassify some of the non-standardized emotional facial expressions as neutral. In sum, emotion recognition by automated facial expression recognition can be an attractive alternative to human emotion recognition for standardized and non-standardized emotional facial expressions. However, we also found limitations in accuracy for specific facial expressions; clearly there is need for thorough empirical evaluation to guide future developments in computer vision of emotional facial expressions.
| S-EPMC8131548 | biostudies-literature
Project description:The present study investigated (1) how well humans can recognize facial expressions represented by a small set of landmarks, a commonly used technique in facial recognition in machine learning and (2) differences in conscious observational behaviors to recognized different types of expressions. Our video stimuli consisted of facial expression represented by 68 landmark points. Conscious observational behaviors were measured by movements of the mouse cursor where a small area around it was only visible to participants. We constructed Bayesian models to analyze how personality traits and observational behaviors influenced how participants recognized different facial expressions. We found that humans could recognize positive expressions with high accuracy, similar to machine learning, even when faces were represented by a small set of landmarks. Although humans fared better than machine learning, recognition of negative expressions was not as high as positives. Our results also showed that personality traits and conscious observational behaviors significantly influenced recognizing facial expressions. For example, people with high agreeableness could correctly recognize faces expressing happiness by observing several areas among faces without focusing on any specific part for very long. These results suggest a mechanism whereby personality traits lead to different conscious observational behaviors and recognitions of facial expressions are based on information obtained through those observational behaviors.
| S-EPMC10550163 | biostudies-literature
Project description:Experimental research examining emotional processes is typically based on the observation of images with affective content, including facial expressions. Future studies will benefit from databases with emotion-inducing stimuli in which characteristics of the stimuli potentially influencing results can be controlled. This study presents Portuguese normative data for the identification of seven facial expressions of emotions (plus a neutral face), on the Radboud Faces Database (RaFD). The effect of participants' gender and models' sex on emotion recognition was also examined. Participants (N = 1249) were exposed to 312 pictures of white adults displaying emotional and neutral faces with a frontal gaze. Recognition agreement between the displayed and participants' chosen expressions ranged from 69% (for anger) to 97% (for happiness). Recognition levels were significantly higher among women than among men only for anger and contempt. The emotion recognition was higher either in female models or in male models depending on the emotion. Overall, the results show high recognition levels of the facial expressions presented, indicating that the RaFD provides adequate stimuli for studies examining the recognition of facial expressions of emotion among college students. Participants' gender had a limited influence on emotion recognition, but the sex of the model requires additional consideration.
| S-EPMC7599941 | biostudies-literature
Project description:Our reactions to facial self-resemblance could reflect either specialized responses to cues of kinship or by-products of the general perceptual mechanisms of face encoding and mere exposure. The adaptive hypothesis predicts differences in reactions to self-resemblance in mating and prosocial contexts, while the by-product hypothesis does not. Using face images that were digitally transformed to resemble participants, I showed that the effects of resemblance on attractiveness judgements depended on both the sex of the judge and the sex of the face being judged: facial resemblance increased attractiveness judgements of same-sex faces more than other-sex faces, despite the use of identical procedures to manipulate resemblance. A control experiment indicated these effects were caused neither by lower resemblance of other-sex faces than same-sex faces, nor by an increased perception of averageness or familiarity of same-sex faces due to prototyping or mere exposure affecting only same-sex faces. The differential impact of self-resemblance on our perception of same-sex and other-sex faces supports the hypothesis that humans use facial resemblance as a cue of kinship.
| S-EPMC1691833 | biostudies-literature
Project description:Not available
2009-12-18 | PRD000081 | Pride
Project description:The ability to recognise facial emotions is essential for successful social interaction. The most common stimuli used when evaluating this ability are photographs. Although these stimuli have proved to be valid, they do not offer the level of realism that virtual humans have achieved. The objective of the present paper is the validation of a new set of dynamic virtual faces (DVFs) that mimic the six basic emotions plus the neutral expression. The faces are prepared to be observed with low and high dynamism, and from front and side views. For this purpose, 204 healthy participants, stratified by gender, age and education level, were recruited for assessing their facial affect recognition with the set of DVFs. The accuracy in responses was compared with the already validated Penn Emotion Recognition Test (ER-40). The results showed that DVFs were as valid as standardised natural faces for accurately recreating human-like facial expressions. The overall accuracy in the identification of emotions was higher for the DVFs (88.25%) than for the ER-40 faces (82.60%). The percentage of hits of each DVF emotion was high, especially for neutral expression and happiness emotion. No statistically significant differences were discovered regarding gender. Nor were significant differences found between younger adults and adults over 60 years. Moreover, there is an increase of hits for avatar faces showing a greater dynamism, as well as front views of the DVFs compared to their profile presentations. DVFs are as valid as standardised natural faces for accurately recreating human-like facial expressions of emotions.
| S-EPMC7833130 | biostudies-literature
Project description:DamID is a powerful technique for identifying regions of the genome bound by a DNA-binding (or DNA-associated) protein. Currently no method exists for automatically processing next-generation sequencing DamID (DamID-seq) data, and the use of DamID-seq datasets with normalisation based on read-counts alone can lead to high background and the loss of bound signal. DamID-seq thus presents novel challenges in terms of normalisation and background minimisation. We describe here damidseq_pipeline, a software pipeline that performs automatic normalisation and background reduction on multiple DamID-seq FASTQ or BAM datasets.
2015-06-26 | GSE69184 | GEO