Work

Design and Implementation of an Intelligent Emotional Analysis and Healing System for University Students

Year: 2025

Type: conference-paper

Abstract: This study presents the design and implementation of a multimodal emotional support system tailored for university students, integrating deep learning-based emotion recognition with AIGC-based artistic generation. The system analyzes uploaded facial images to recognize emotions, constructs an anxiety scoring model, and maps the results to controllable prompts. These prompts guide the generation of therapeutic images and videos, aiming to produce emotionally suggestive and aesthetically healing content. Image generation is powered by the Kolors model, while video generation is based on Tongyi Wanxiang, with the overall process coordinated by the Qwen language model. Experimental results demonstrate that the system accurately identifies students’ anxiety states and provides visual channels for emotional expression and release through generative content, offering practical value in psychological support.

Source: International Conference on Communication and Information Processing (ICCIP)

DOI: 10.1145/3784833.3784858

Authors: Xiaofa Zheng, FuMo Luo, Jie Chang, Yunpeng Jia, Jun Chen

Institutions: Beijing University of Posts and Telecommunications, Beijing Film Academy, Beijing Jingbei Vocational College

Language: English

Topic: Emotion and Mood Recognition

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Quality education

Open Access status: hybrid

Funder: National Office for Philosophy and Social Sciences