DEFAEK: Domain Effective Fast Adaptive Network for Face Anti-Spoofing.

Existing deep learning based face anti-spoofing (FAS) or deepfake detection approaches usually rely on large-scale datasets and powerful networks with significant amount of parameters to achieve satisfactory performance. However, these make them resource-heavy and unsuitable for handheld devices. Moreover, they are limited by the types of spoof in the dataset they train on and require considerable training time. To produce a robust FAS model, they need large datasets covering the widest variety of predefined presentation attacks possible. Testing on new or unseen attacks or environments generally results in poor performance. Ideally, the FAS model should learn discriminative features that can generalize well even on unseen spoof types. In this paper, we propose a fast learning approach called Domain Effective Fast Adaptive nEt-worK (DEFAEK), a face anti-spoofing approach based on the optimization-based meta-learning paradigm that effectively and quickly adapts to new tasks. DEFAEK treats differences in an environment as domains and simulates multiple domain shifts during training. To further improve the effectiveness and efficiency of meta-learning, we adopt the metric learning in the inner loop update with careful sample selection. With extensive experiments on the challenging CelebA-Spoof and FaceForensics++ datasets, the evaluation results show that DEFAEK can learn cues independent of the environment with good generalization capability. In addition, the resulting model is lightweight following the design principle of modern lightweight network architecture and still generalizes well on unseen classes. In addition, we also demonstrate our model's capabilities by comparing the numbers of parameters, FLOPS, and model performance with other state-of-the-art methods.

The effect of diagnostic labels on teachers' perceptions of behaviours of students with autism spectrum disorder.

BACKGROUND: In Singapore, despite the availability of specialized services and resources in mainstream schools for students with special educational needs, parents can still be hesitant to disclose their child's diagnosis of autism spectrum disorder (ASD), for fear of stigma or negative attitudes from teachers and peers. AIM: This study aims to understand how diagnostic labels affect teachers' perceptions of behaviours in students with ASD in mainstream primary schools, particularly behaviours which are difficult for educators to manage in the classroom setting. Based on research suggesting that a diagnostic label can provide an alternative explanation for atypical behaviours and reduce negative impressions, we hypothesize that teachers would perceive these behaviours of children with ASD less negatively as compared to children without any diagnostic labels. SAMPLE: The sample consisted of 120 mainstream primary school teachers who reported a mean of 9.97 years (SD = 7.96) of teaching experience. Majority of participants (65%) indicated that they had direct experience of teaching students with ASD in mainstream schools (mean years = 3.58, SD = 4.77). METHOD: Participants read a total of 20 vignettes depicting a range of behaviours typically observed in primary school-aged students with ASD. Participants were asked to rate the featured student's behaviour on a 5-point Likert scale (i.e., 1 = strongly negative, 2 = negative, 3 = neutral, 4 = positive, and 5 = strongly positive). The survey vignettes were presented in a randomized order, and participants were randomly assigned to either the experimental group (awareness of diagnostic label) or the control group (non-awareness of label). RESULTS: Results indicated that the experimental group rated the behaviours less negatively than the control group. Qualitative feedback from participants also indicated that knowledge of the diagnostic label helped them to perceive the behaviours more positively. CONCLUSION: Findings from this study have implications for parents regarding diagnosis disclosure and may serve as an encouragement to parents to consider disclosing their child's ASD diagnosis to the school.

Anxiety symptoms in young people with autism spectrum disorder attending special schools: Associations with gender, adaptive functioning and autism symptomatology.

Anxiety-related problems are among the most frequently reported mental health difficulties in autism spectrum disorder. As most research has focused on clinical samples or high-functioning children with autism spectrum disorder, less is known about the factors associated with anxiety in community samples across the ability range. This cross-sectional study examined the association of gender, age, adaptive functioning and autism symptom severity with different caregiver-reported anxiety symptoms. Participants were caregivers of 241 children (6-18 years old) with autism spectrum disorder attending special schools in Singapore. Measures included the Spence Children's Anxiety Scale and assessments of overall emotional, behavioural and adaptive functioning. Caregivers reported more anxiety symptoms in total, but fewer social anxiety symptoms, than Spence Children's Anxiety Scale Australian/Dutch norms. There were no gender differences. Variance in total anxiety scores was best explained by severity of repetitive speech/stereotyped behaviour symptoms, followed by adaptive functioning. Severity of repetitive speech/behaviour symptoms was a significant predictor of separation anxiety, generalized anxiety, panic/agoraphobia and obsessive-compulsive subscale symptoms, but not of social phobia and physical injury fears. Adaptive functioning and chronological age predicted social phobia and generalized anxiety symptoms only. Severity of social/communication autism symptoms did not explain any anxiety symptoms, when the other variables were controlled for. Findings are discussed in relation to the existing literature. Limitations and possible implications for prevention, assessment and intervention are also discussed.

A preliminary investigation of the Spence Children's Anxiety Parent Scale as a screening tool for anxiety in young people with autism spectrum disorders.

Despite high rates of clinically elevated anxiety difficulties in children and adolescents with autism spectrum disorders (ASDs), very few studies have systematically examined the usefulness of commonly used caregiver report anxiety screening tools with this population. This study investigated the use of the Spence Children's Anxiety Scale-Parent version (SCAS-P) as a screening tool for anxiety disorders when compared to a standardized DSM-IV-TR-based clinical interview, the Kiddie-Schedule for Schizophrenia and Affective Disorders-Present and Lifetime version (K-SADS-PL). Thirty-two caregivers of youth with a clinical diagnosis of ASD (mean age 10.3 years) attending a specialist autism school participated in this study. They first completed the SCAS-P, a measure of adaptive functioning and a checklist of other emotional and behavioral difficulties. They were then interviewed with the K-SADS-PL. Internal consistency for the SCAS Total score was .88, but Cronbach's alphas were <.70 in three of the six SCAS-P subscales. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the SCAS-P against K-SADS diagnosis were .75, .71, .27, and .95, respectively. All values were >.70, except for the PPV. Evidence of convergent validity between the SCAS-P, K-SADS-PL and DBC anxiety subscale was also found. The high false positive rates notwithstanding, the preliminary data of acceptable to excellent sensitivity, specificity and NPV values tentatively suggest that the SCAS-P may be useful for screening non-help seeking young people with ASD for elevated anxiety symptoms. Further replication in larger studies is needed and ways in which the SCAS-P could be further developed and investigated for use with youth with ASD are discussed.