MPDD-AVG @ ACM MM 2026 advances personalized depression detection by integrating audio-visual, gait, and personality data across young and elderly populations.
Depression is a prevalent mental health disorder affecting individuals across the lifespan, with significant impact on young adults and the elderly population. However, existing depression detection approaches predominantly rely on conversational or interview-based modalities with limited age diversity, while ambulatory behavioral signatures such as gait characteristics — though recognized as important clinical indicators of psychomotor symptoms — remain largely unexplored.
Moreover, current methods establish direct data-to-score mappings without modeling individual differences, overlooking psychomotor domains and the inter-individual heterogeneity attributable to personality profiles, demographic variables, and comorbid conditions.
To address these limitations, we introduce MPDD-AVG, a comprehensive benchmark that uniquely integrates two activities of semi-structured interview behavioral data and gait monitoring from wearable sensors. The challenge is an updated version of MPDD2025 @ ACM MM 2025, which comprises two age-specific datasets: MPDD-Young (110 young adults) investigating academic stress and social functioning, and MPDD-Elderly (110 older adults) examining late-life depression influenced by chronic conditions and living arrangements.
Critically, we provide raw individual difference annotations including Big Five-10 personality dimensions, demographic variables, and health conditions — rather than pre-engineered features only — explicitly encouraging participants to develop innovative personality-conditioned modeling strategies.
Semi-structured interviews capturing cognitive-linguistic and affective-paralinguistic behavioral cues from both young adults and elderly participants.
Wearable IMU sensors capturing walking speed, stride regularity, and postural dynamics as psychomotor indicators of depression.
Raw Big Five-10 personality annotations with demographic and health condition data to model inter-individual heterogeneity in depressive symptomatology.
110 young adults (academic stress, social functioning) and 110 elderly participants (late-life depression, chronic conditions, living arrangements).
PHQ-9 depression scores, Big Five-10 personality traits, demographics, disease labels — the most complete annotation scheme to date.
Both classification (binary/ternary/quinary) and regression tasks with composite metrics covering Macro-F1, CCC, and Cohen's Kappa.