Summary of Evaluating Image-based Face and Eye Tracking with Event Cameras, by Khadija Iddrisu et al.
Evaluating Image-Based Face and Eye Tracking with Event Cameras
by Khadija Iddrisu, Waseem Shariff, Noel E.OConnor, Joseph Lemley, Suzanne Little
First submitted to arxiv on: 19 Aug 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 showcases the viability of integrating conventional algorithms with event-based data from Neuromorphic sensors, also known as Event Cameras. These cameras capture changes in local light intensity at the pixel level, producing asynchronously generated data termed “events”. The researchers demonstrate face and eye tracking using a frame format transformed from event data, leveraging GR-YOLO, a technique derived from YOLOv3. They compare this approach to training with YOLOv8 on the publicly accessible Helen Dataset and evaluate the models’ performance on real event streams from Prophesee’s event cameras and the Faces in Event Stream (FES) benchmark dataset. The results show a mean Average precision score of 0.91, demonstrating robust performance under varying light conditions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Event cameras capture changes in local light intensity at the pixel level, producing asynchronously generated data termed “events”. This study demonstrates face and eye tracking using event-based data from Neuromorphic sensors. The researchers transform event data into a frame format and use GR-YOLO, a technique derived from YOLOv3, for face and eye detection tasks. They compare this approach to training with YOLOv8 on the Helen Dataset. The study evaluates the models’ performance on real event camera data under varying light conditions. |
Keywords
» Artificial intelligence » Mean average precision » Tracking » Yolo
