Summary of Ic3m: In-car Multimodal Multi-object Monitoring For Abnormal Status Of Both Driver and Passengers, by Zihan Fang et al.
IC3M: In-Car Multimodal Multi-object Monitoring for Abnormal Status of Both Driver and Passengers
by Zihan Fang, Zheng Lin, Senkang Hu, Hangcheng Cao, Yiqin Deng, Xianhao Chen, Yuguang Fang
First submitted to arxiv on: 3 Oct 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Systems and Control (eess.SY)
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 The proposed IC3M framework is an efficient camera-rotation-based multimodal system designed to monitor both the driver and passengers in a car. It tackles the challenges of class imbalance, missing modalities, and limited labeled data by introducing two key modules: an adaptive threshold pseudo-labeling strategy and a missing modality reconstruction. The first module customizes pseudo-labeling thresholds for different classes based on the class distribution, generating class-balanced pseudo labels to guide model training effectively. The second module leverages crossmodality relationships learned from limited labels to accurately recover missing modalities by distribution transferring from available modalities. Experimental results demonstrate that IC3M outperforms state-of-the-art benchmarks in accuracy, precision, and recall while exhibiting superior robustness under limited labeled data and severe missing modality. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary IC3M is a new way to use cameras to monitor both drivers and passengers in cars. The system tries to solve some big problems like having too many or too few examples of what’s normal versus abnormal behavior. It also deals with situations where certain camera views are missing, which makes things harder. To do this, IC3M has two main parts: one that helps the computer learn from the data it has, and another that can fill in gaps in the camera views by using information from other cameras. |
Keywords
» Artificial intelligence » Precision » Recall
