InCarEmo: A Multimodal Dataset for In-Cabin Emotion Recognition and Driver State Monitoring

arXiv cs.AI · 6 hours ago

Researchers introduced InCarEmo, a multimodal dataset combining video, audio, and text from scripted in-cabin scenarios to train AI systems for emotion recognition and driver state monitoring. The dataset includes RGB and infrared video, audio, and dialogue text in Chinese with an English benchmark, supporting three tasks: emotion recognition, fatigue detection, and distraction monitoring. This enables development of safer in-cabin systems that can detect driver state under various conditions including low-light and noisy environments.