Deep Vision Analysis for Customers' Perception and Service Refinement Under Artificial Intelligence

To improve the tourist experience and service quality in the ice and snow tourism scene, this study discusses tourists' behavioral characteristics and emotional needs based on deep vision analysis using artificial intelligence (AI). Multimodal data fusion technology is introduced, combining facial expressions, body posture, and environmental factors to construct an emotional and behavioral analysis model suitable for ice and snow tourism scenes. The optimized model is compared with Swin Transformer, EfficientNetV2, Time-Space Transformer (TimeSForm), and other models. The results show the optimized model excels in predicting tourists' behavior and analyzing demand characteristics. This study provides a new technical approach for analyzing emotion and behavior in ice and snow tourism scenes and makes a valuable contribution to research in the field of ice and snow tourism.