Zero-Shot Pupil Segmentation with SAM 2: A Case Study of Over 14 Million Images

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Maquiling, Virmarie, Byrne, Sean Anthony, Niehorster, Diederick C., Carminati, Marco, Kasneci, Enkelejda
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929673315811328
author Maquiling, Virmarie
Byrne, Sean Anthony
Niehorster, Diederick C.
Carminati, Marco
Kasneci, Enkelejda
author_facet Maquiling, Virmarie
Byrne, Sean Anthony
Niehorster, Diederick C.
Carminati, Marco
Kasneci, Enkelejda
contents We explore the transformative potential of SAM 2, a vision foundation model, in advancing gaze estimation and eye tracking technologies. By significantly reducing annotation time, lowering technical barriers through its ease of deployment, and enhancing segmentation accuracy, SAM 2 addresses critical challenges faced by researchers and practitioners. Utilizing its zero-shot segmentation capabilities with minimal user input-a single click per video-we tested SAM 2 on over 14 million eye images from diverse datasets, including virtual reality setups and the world's largest unified dataset recorded using wearable eye trackers. Remarkably, in pupil segmentation tasks, SAM 2 matches the performance of domain-specific models trained solely on eye images, achieving competitive mean Intersection over Union (mIoU) scores of up to 93% without fine-tuning. Additionally, we provide our code and segmentation masks for these widely used datasets to promote further research.
format Preprint
id arxiv_https___arxiv_org_abs_2410_08926
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Zero-Shot Pupil Segmentation with SAM 2: A Case Study of Over 14 Million Images
Maquiling, Virmarie
Byrne, Sean Anthony
Niehorster, Diederick C.
Carminati, Marco
Kasneci, Enkelejda
Computer Vision and Pattern Recognition
Artificial Intelligence
Human-Computer Interaction
We explore the transformative potential of SAM 2, a vision foundation model, in advancing gaze estimation and eye tracking technologies. By significantly reducing annotation time, lowering technical barriers through its ease of deployment, and enhancing segmentation accuracy, SAM 2 addresses critical challenges faced by researchers and practitioners. Utilizing its zero-shot segmentation capabilities with minimal user input-a single click per video-we tested SAM 2 on over 14 million eye images from diverse datasets, including virtual reality setups and the world's largest unified dataset recorded using wearable eye trackers. Remarkably, in pupil segmentation tasks, SAM 2 matches the performance of domain-specific models trained solely on eye images, achieving competitive mean Intersection over Union (mIoU) scores of up to 93% without fine-tuning. Additionally, we provide our code and segmentation masks for these widely used datasets to promote further research.
title Zero-Shot Pupil Segmentation with SAM 2: A Case Study of Over 14 Million Images
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Human-Computer Interaction
url https://arxiv.org/abs/2410.08926