LMVD
A dataset for multimodal depression detection derived from video blog scenarios. It contains multiple modalities and is used to evaluate depression risk prediction models.
Timeline (last 90 days)
With a frozen 7B backbone, EDA Support improves F1 by 7.61 and 9.88 percentage points over direct prediction on D-Vlog Leave-Query-Out and LMVD, respectively.
D-Vlog and LMVD are two commonly used multimodal depression detection datasets; D-Vlog is based on video blog self-narratives and contains segment-level emotion and depression-related labels; LMVD targets interview-style scenarios and contains structured question-and-answer dialogue and scale scores.
All Facts (2)
With a frozen 7B backbone, EDA Support improves F1 by 7.61 and 9.88 percentage points over direct prediction on D-Vlog Leave-Query-Out and LMVD, respectively.
50%UnverifiedD-Vlog and LMVD are two commonly used multimodal depression detection datasets; D-Vlog is based on video blog self-narratives and contains segment-level emotion and depression-related labels; LMVD targets interview-style scenarios and contains structured question-and-answer dialogue and scale scores.
50%