Summary of Efficient Mixture-of-expert For Video-based Driver State and Physiological Multi-task Estimation in Conditional Autonomous Driving, by Jiyao Wang et al.
Efficient Mixture-of-Expert for Video-based Driver State and Physiological Multi-task Estimation in Conditional Autonomous Driving
by Jiyao Wang, Xiao Yang, Zhenyu Wang, Ximeng Wei, Ange Wang, Dengbo He, Kaishun Wu
First submitted to arxiv on: 28 Oct 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This study proposes a novel multi-task Driver Monitoring System (DMS) called VDMoE that leverages RGB video input to monitor driver states non-invasively in SAE Level-2/3 autonomous driving contexts. The DMS utilizes key facial features to minimize computational load and integrates remote Photoplethysmography (rPPG) for physiological insights, enhancing detection accuracy while maintaining efficiency. The approach optimizes the Mixture-of-Experts (MoE) framework to accommodate multi-modal inputs and improves performance across different tasks. A novel prior-inclusive regularization method is introduced to align model outputs with statistical priors, accelerating convergence and mitigating overfitting risks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper explores ways to make self-driving cars safer by creating a system that monitors drivers while they’re using the car. The current problem is that if humans are doing other tasks while driving, it can be dangerous because they might not notice what’s happening on the road. To solve this issue, the researchers developed a new way to track drivers’ cognitive load and drowsiness using cameras and sensors. This system is designed for autonomous cars with Level-2/3 automation, which means humans are still involved in driving but can take their eyes off the road sometimes. |
Keywords
» Artificial intelligence » Mixture of experts » Multi modal » Multi task » Overfitting » Regularization
