Reproduce “Artificial Intelligence Methods in In-Cabin Use Cases: A Survey” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yao Rong, Chao Han, Christian Hellert, Antje Loyal, Enkelejda Kasneci
This survey provides a comprehensive review of Artificial Intelligence methods applied to in-cabin use cases within autonomous driving scenarios, categorizing applications by driving safety and comfort. It identifies prevalent AI algorithms and input features, demonstrating that machine learning and deep learning dominate in-cabin tasks while revealing how use-case relevance shifts across SAE automation levels.
Key insights
- Machine Learning (50.0%) and Deep Learning (40.5%) are the overwhelmingly dominant AI methodologies employed across in-cabin use-cases, primarily addressing classification and regression problems.
- Key input features like driver's eye/head movements, full driver/road images, and vehicle dynamics are crucial across numerous applications for both safety and comfort.
- The purpose and relevance of in-cabin AI functions evolve with increasing SAE automation levels; for instance, driver anomaly detection in L3 transitions to passenger emotion and satisfaction monitoring in L4/L5.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.