LMVD: A Large-Scale Multimodal Vlog Dataset for Depression Detection in the Wild

• DOI: 10.36227/techrxiv.171591570.08868181/v1
• Publication Date: 2024-05-17T03:15:13Z
• Authors: Lang He; Kai Chen; Junnan Zhao; Yimeng Wang; Ercheng Pei; Haifeng Chen; Jiewei Jiang; Shiqing Zhang; Jie Zhang; Zhongmin Wang; Tao He; Prayag Tiwari
• Keywords: Depression

Depression can significantly impact many aspects of an individual's life, including their personal and social functioning, academic work performance, overall quality life.Many researchers within the field affective computing are adopting deep learning technology to explore potential patterns related detection depression.However, because subjects' privacy protection concerns, that data in this area is still scarce, presenting a challenge for discriminative models used detecting depression.To navigate these obstacles, large-scale multimodal vlog dataset (LMVD), depression recognition wild built.In LMVD, which has 1823 samples with 214 hours 1475 participants captured from four multimedia platforms (Sina Weibo, Bilibili, Tiktok, YouTube).A novel architecture termed MDDformer learn non-verbal behaviors individuals proposed.Extensive validations performed on LMVD dataset, demonstrating superior performance detection.We anticipate will contribute valuable function community.The code released at link: