In this study, we propose and experimentally validate a multisource-sensor-fusion (MSF)-based immersive interactive projection system (IIPS) integrating a depth camera, a distance-sensing module, and an audio-input module. The system is designed to enhance interaction recognition accuracy, response consistency, and robustness in dynamic environments, where single-sensor approaches are often affected by occlusion, environmental noise, and event misclassification. A sensor fusion framework is employed to integrate heterogeneous sensing signals at the decision level for real-time interaction detection and event triggering. By leveraging complementary sensing modalities, the system has an improved environmental perception capability and interaction reliability, enabling stable and adaptive user–system interaction in complex exhibition scenarios. Two complementary evaluations were conducted. First, system-level performance was quantitatively assessed using interaction recognition accuracy, response latency, trigger success rate, and operational stability. The proposed system achieved 93.4% recognition accuracy, 186 ms average response latency, 91.8% trigger success, and only two abnormal events over 30 continuous test rounds, demonstrating high robustness and real-time performance. Second, user evaluation data from 384 participants were analyzed using exploratory factor analysis, confirmatory factor analysis, and structural equation modeling. Unlike conventional IIPS studies that primarily focus on visual rendering or isolated interaction performance, in this study, we emphasize the integration of multisource sensing reliability, system interaction performance, and user-experience evaluation within a unified analytical framework. The primary objective of this study is to investigate how MSF affects both technical interaction performance and experiential outcomes in immersive environments. The main contribution lies in establishing a sensor-oriented evaluation framework that directly links sensing quality with immersion, satisfaction, and revisit intention through empirical structural modeling. However, the current framework is limited to a specific experimental environment and a fixed combination of sensing modules, which may affect the generalizability of the results. Overall, the results demonstrate that MSF is a critical design strategy for improving sensing reliability, interaction performance, and experiential quality, providing practical insights for the development of sensor-integrated IIPS.