The paper introduces PriMonitor, an adaptive privacy-preserving approach for multimodal emotion detection in driver monitoring systems (DMS) within the context of edge computing and the Internet of Vehicles (IoV). The authors address the growing concern of privacy data collection from in-vehicle cameras and microphones, which can lead to privacy re-identification when data from multiple modalities are correlated. PriMonitor employs a generalized random response-based differential privacy method to enhance text privacy protection and ensure privacy across multiple modalities. The approach includes pre-aggregator and iterative mechanisms to determine suitable weight assignments within a given privacy budget. Experimental results demonstrate the efficiency and competitiveness of PriMonitor in maintaining high accuracy while effectively mitigating privacy re-identification. The paper highlights the importance of multimodal emotion detection in enhancing driving safety and the need for robust privacy-preserving techniques to protect driver privacy.The paper introduces PriMonitor, an adaptive privacy-preserving approach for multimodal emotion detection in driver monitoring systems (DMS) within the context of edge computing and the Internet of Vehicles (IoV). The authors address the growing concern of privacy data collection from in-vehicle cameras and microphones, which can lead to privacy re-identification when data from multiple modalities are correlated. PriMonitor employs a generalized random response-based differential privacy method to enhance text privacy protection and ensure privacy across multiple modalities. The approach includes pre-aggregator and iterative mechanisms to determine suitable weight assignments within a given privacy budget. Experimental results demonstrate the efficiency and competitiveness of PriMonitor in maintaining high accuracy while effectively mitigating privacy re-identification. The paper highlights the importance of multimodal emotion detection in enhancing driving safety and the need for robust privacy-preserving techniques to protect driver privacy.