Depression tendency evaluation system and method based on multi-modal characteristics

The invention discloses a depression tendency evaluation system and method based on multi-modal features. The depression tendency evaluation system comprises an original data acquisition device, a data preprocessing device, a multi-modal feature extraction device and a result output device. The original data acquisition device is used for acquiring original data, the data preprocessing device is used for extracting to-be-processed data from the original data, the multi-modal feature extraction device is used for acquiring multi-modal feature data according to the to-be-processed data, and the result output device is used for inputting the multi-modal feature data into a depression tendency evaluation mathematical model. And obtaining a depression tendency evaluation result. The multi-modal feature data comprises the facial feature data, the audio feature data and the gait feature data of the to-be-tested user, and the multi-modal features are input into the depression tendency evaluation mathematical model to