Audiovisual foundation
CompletedLocked engineering evaluation on public audiovisual data; calibration, robustness, OOD and runtime measured.
Evidence library · External scientific context
Make every claim traceable.
NDMS sits across multimodal affective computing, audiovisual fusion, uncertainty-aware ML, behavioral signal analysis, autistic masking research and responsible AI. This library links the scientific context directly rather than asking visitors to take the architecture on trust.
These papers support the scientific context, measurement choices, benchmark conventions and risks relevant to NDMS. Unless a source explicitly evaluates NDMS, it is not presented as external validation of NDMS. Independent replication remains a future milestone.
Forward path
What exists now—and what has to happen next.
This is the confidence map for NDMS. Each step has a different evidence standard; later stages are not implied by earlier engineering results.
Prototype instrumentation
ActiveProduct surfaces now expose signal quality, uncertainty and engineering state rather than hiding model confidence.
Dedicated consented dataset
NextCollect fit-for-purpose longitudinal audio/video plus contextual self-report under explicit consent and actor-independent protocols.
Prospective masking validation
PlannedPre-register constructs, labels, endpoints and failure criteria before testing masking-specific hypotheses.
External replication
TargetIndependent or multi-site replication with cross-dataset and subgroup analyses before stronger generalization claims.
Regulated / commercial pathways
ConditionalAny clinical claims remain separate and would require dedicated evidence, governance and regulatory review.
Illustrative figures + real research imagery
Architecture should be understandable without pretending a diagram is data.
Illustrative system figures are explicitly separated from experimental plots. Real stock photographs are used only as editorial context, with source attribution.
Illustrative figureNDMS evidence stack
SignalsAudio · video · context
→FusionQuality-aware temporal modeling
→AssuranceCalibration · OOD · uncertainty
→EvidenceLocked test · stress tests · prospective studies
Research library
55 linked papers, reviews and benchmark sources.
Every source opens to a DOI, publisher, proceedings page or preprint record. Filters make the library useful to researchers rather than decorative SEO text.
Showing 55 of 55
R1R2R3R4R5R6R7R8R9R10R11R12R13R14R15R16R17R18R19R20R21R22R23R24R25R26R27R28R29R30R31R32R33R34R35R36R37R38R39R40R41R42R43R44R45R46R47R48R49R50R51R52R53R54R55
CREMA-D: Crowd-Sourced Emotional Multimodal Actors Dataset
IEEE Transactions on Affective Computing
Foundational audiovisual emotion corpus used in the current NDMS engineering evaluation.The Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS)
PLOS ONE
Widely used controlled audiovisual emotion benchmark.IEMOCAP: Interactive emotional dyadic motion capture database
Language Resources and Evaluation
Dyadic multimodal emotion corpus with audio, video and motion capture.Multimodal Language Analysis in the Wild: CMU-MOSEI Dataset and Interpretable Dynamic Fusion Graph
ACL
Large-scale multimodal benchmark for in-the-wild affective and sentiment modeling.RECOLA: Remote Collaborative and Affective Interactions
IEEE FG
Continuous multimodal affect dataset used in temporal emotion research.Aff-Wild: Valence and Arousal 'In-the-Wild' Challenge
CVPR Workshops
In-the-wild facial affect benchmark emphasizing temporal variation.MSP-IMPROV: An Acted Corpus of Dyadic Interactions to Study Emotion Perception
IEEE Transactions on Affective Computing
Naturalistic dyadic corpus used in audiovisual affect research.EmoWork: A Multimodal Dataset for Assessing Emotion, Stress, and Emotional Workload in Interpersonal Work Scenario
Scientific Data
Context-rich multimodal dataset combining behavioral, physiological and self-report signals.Continuous Prediction of Spontaneous Affect from Multiple Cues and Modalities in Valence-Arousal Space
IEEE Transactions on Affective Computing
Classic multimodal temporal affect modeling across facial, gesture and audio cues.AuxFormer: Robust Approach to Audiovisual Emotion Recognition
ICASSP
Transformer-based audiovisual fusion with robustness to missing modalities.Robust Audiovisual Emotion Recognition: Aligning Modalities, Capturing Temporal Information, and Handling Missing Features
IEEE Transactions on Affective Computing
Directly relevant to alignment, temporal modeling and missing-modality robustness.Audio-Visual Fusion for Emotion Recognition in the Valence-Arousal Space Using Joint Cross-Attention
IEEE Transactions on Biometrics, Behavior, and Identity Science
Cross-attention fusion designed for noisy or absent modalities.Modelling Stochastic Context of Audio-Visual Expressive Behaviour with Affective Processes
IEEE Transactions on Affective Computing
Probabilistic temporal context modeling for audiovisual affect.Robust Audio-Visual Fusion for Emotion Recognition Based on Cross-Modal Learning under Noisy Conditions
Computers, Materials & Continua
Cross-modal learning under environmental noise and visual obstruction.Multimodal Emotion Recognition using Audio-Video Transformer Fusion with Cross Attention
arXiv
Transformer cross-attention evaluated across several common audiovisual datasets.LaWNet: Audio-Visual Emotion Recognition by Listening and Watching
IJCNN
Cross-attention audiovisual fusion evaluated on RAVDESS and CREMA-D.AAT-CGF: A Cross-Modal Deep Fusion Framework with Attention Aggregation and Cross Graph Fusion for Multimodal Emotion Recognition
IALP
Graph-based cross-modal fusion across audio, visual and text.Multimodal emotion recognition based on a fusion of audiovisual information with temporal dynamics
Multimedia Tools and Applications
Temporal audiovisual fusion across multiple public datasets.COLD Fusion: Calibrated and Ordinal Latent Distribution Fusion for Uncertainty-Aware Multimodal Emotion Recognition
arXiv
Uncertainty-aware audiovisual fusion with explicit calibration objectives.On Calibration of Modern Neural Networks
ICML
Foundational work showing modern neural networks can be miscalibrated and introducing temperature scaling.Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles
NeurIPS
Influential practical baseline for predictive uncertainty.Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
ICML
Widely used approximate Bayesian approach for predictive uncertainty.A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
ICLR
Foundational confidence-based OOD detection baseline.Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks
ICLR
Introduces ODIN for stronger OOD separation.Energy-based Out-of-distribution Detection
NeurIPS
Energy score approach to identifying distribution shift.Revisiting One-vs-All Classifiers for Predictive Uncertainty and Out-of-Distribution Detection in Neural Networks
arXiv
Studies classifier parameterization effects on uncertainty and OOD behavior.Towards trustworthy predictions from deep neural networks with fast adversarial calibration
AAAI
Calibration under domain shift using adversarial training objectives.A Survey of Uncertainty in Deep Neural Networks
Artificial Intelligence Review
Broad survey of epistemic/aleatoric uncertainty, calibration and practical limitations.Density Uncertainty Layers for Reliable Uncertainty Estimation
arXiv
Links predictive uncertainty to empirical input density.Amortized Conditional Normalized Maximum Likelihood: Reliable Out of Distribution Uncertainty Estimation
arXiv
General uncertainty and calibration method under distribution shift.Recognising spontaneous facial micro-expressions
ICCV
Early spontaneous micro-expression corpus and recognition work.Video-based Facial Micro-Expression Analysis: A Survey of Datasets, Features and Algorithms
IEEE TPAMI
Comprehensive survey of datasets, features, evaluation protocols and limitations.Spontaneous Facial Micro-expression Recognition via Deep Convolutional Network
IPTA
Deep temporal modeling evaluated on SMIC, CASME and CASME2.Effective DemeapexNet: Revealing Spontaneous Facial Micro-Expressions
ICCDA
Apex-frame micro-expression recognition with transfer learning.A Review of 25 Spontaneous and Dynamic Facial Expression Databases of Basic Emotions
Affective Science
Reviews ecological validity, technical properties and ethical constraints of facial-expression datasets.Advances in Facial Micro-Expression Detection and Recognition: A Comprehensive Review
Information
Recent review of detection, recognition, datasets and generalization challenges.MMED: A Multimodal Micro-Expression Dataset based on Audio-Visual Fusion
arXiv
Recent multimodal micro-expression dataset incorporating vocal cues and LOSO evaluation.Emotion Recognition Using Different Sensors, Emotion Models, Methods and Datasets: A Comprehensive Review
Sensors
Compares behavioral and physiological sensors across emotion-recognition systems.Multimodal emotion recognition: A comprehensive review, trends, and challenges
WIREs Data Mining and Knowledge Discovery
Large review of multimodal emotion theory, datasets, fusion strategies and open challenges.A Systematic Review on Multimodal Emotion Recognition: Building Blocks, Current State, Applications, and Challenges
IEEE Access
Systematic review of multimodal emotion-recognition methods and research gaps.“Putting on My Best Normal”: Social Camouflaging in Adults with Autism Spectrum Conditions
Journal of Autism and Developmental Disorders
Foundational qualitative study on motivations, strategies and costs of camouflaging.Development and Validation of the Camouflaging Autistic Traits Questionnaire (CAT-Q)
Journal of Autism and Developmental Disorders
Validated self-report measurement framework for social camouflaging.Social camouflaging in autism: Is it time to lose the mask?
Autism
Editorial synthesis of the emerging social-camouflaging literature.Camouflage and masking behavior in adult autism
Frontiers in Psychiatry
Systematic review of motivations, correlates and mental-health impacts.It is More Anxiousness than Role-playing: Social Camouflaging Conceptualization Among Adults on the Autism Spectrum Compared to Persons with Social Anxiety Disorder
Journal of Autism and Developmental Disorders
Examines camouflaging, social anxiety and burnout across comparative groups.What We Know and Do Not Know About Camouflaging, Impression Management, and Mental Health and Wellbeing in Autistic People
Autism Research
Highlights inconsistent evidence and the need to measure effort, context and effectiveness separately.On the Status and Knowledge of Camouflaging, Masking, and Compensatory Behaviors in Autism Spectrum Disorder
Education and Training in Autism and Developmental Disabilities
Reviews the state of evidence around camouflaging, masking and compensation.Masking in autism. Analysis of the phenomenon
Szkoła Specjalna
Narrative review emphasizing masking effort, overload and identity consequences.Camouflage in autism: a bibliometric perspective
Advances in Autism
Maps the scientific literature on autism camouflaging and its dominant themes.Gender differences in self-reported camouflaging in autistic and non-autistic adults
Autism
Examines sex/gender differences in self-reported camouflaging.Is social camouflaging associated with anxiety and depression in autistic adults?
Molecular Autism
Tests relationships between camouflaging and anxiety/depression.Understanding the Relationships Between Autistic Identity, Disclosure, and Camouflaging
Autism in Adulthood
Links camouflaging with identity and disclosure.A Systematic Literature Review of Human-Centered, Ethical, and Responsible AI
arXiv / HCI review
Maps governance, fairness, explainability, privacy and human flourishing in responsible AI.Affective Computing and Emotional Data: Challenges and Implications in Privacy Regulations, The AI Act, and Ethics in Large Language Models
arXiv
Examines consent, emotional data, privacy and regulatory implications.Reimagining Intelligence: A Comprehensive Review of Human-Centric AI Systems and Their Societal and Healthcare Integration
Mesopotamian Journal of AI in Healthcare
Human-centric AI review emphasizing transparency, fairness, autonomy and privacy.Evidence, not decoration
The next confidence jump comes from new data—not more adjectives.
The literature establishes context. NDMS-specific confidence has to come from prospective, actor-independent, longitudinal validation and eventually external replication.