Hand Gesture based driver-vehicle interface Framework for automotive In-Vehicle Infotainment system

The automotive in-vehicle infotainment system is an integral part of the four wheels these days. Driver vehicle interface becomes critical in the vehicle infotainment system. Therefore, it is necessary to develop an intelligent framework to reduce human and physical vehicle interaction. Hand gestures were used as an input modality to reduce visual and cognitive demands. This study has proposed a Single Shot multi-box Detector (SSD) based hand gesture vehicle interface framework to interact with a driver without any eye contact. The SSD model is used to detect hand gestures. Convolution Neural Network (CNN) is used to classify the gesture and make the right decision to control the in-vehicle infotainment system. The five independent hand gesture-based vehicle infotainment control is considered to reduce physical interaction and improve driver safety. To test and validate the proposed framework, a virtual experiment testbed is created using the Unreal Engine. The outcomes demonstrate the efficacy of the proposed work.