COCO Search-18

COCO-Search18 is a high-quality dataset of eye gaze behavior during goal-directed search in natural-scene images. It includes eye movements from 10 participants searching for 18 target-object categories, yielding approximately 300,000 search fixations. COCO-Search18 is part of the MIT/Tuebingen Saliency Benchmark. You can download the training and validation search fixations for COCO-Search18 from here.

If you use COCO-Search18, please cite:

• Chen, Y., Yang, Z., Ahn, S., Samaras, D., Hoai, M., & Zelinsky, G. (2021). COCO-Search18 Fixation Dataset for Predicting Goal-directed Attention Control. Scientific Reports, 11 (1), 1-11, 2021.
• Yang, Z., Huang, L., Chen, Y., Wei, Z., Ahn, S., Zelinsky, G., Samaras, D., & Hoai, M. (2020). Predicting Goal-directed Human Attention Using Inverse Reinforcement Learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 193-202)

SB-SAT

SB-SAT is a high-quality eye gaze dataset of reading behavior that includes eye movements recorded from 95 individuals reading four published SAT passages. Each passage is followed by corresponding SAT questions and self-evaluation questionnaires that measure engagement, sleepiness, and other factors. You can download all recorded fixation and response data from here.

If you use SB-SAT, please cite:

• Ahn, S., Kelton, C., Balasubramanian, A., & Zelinsky, G. (2020, June). Towards predicting reading comprehension from gaze behavior. In ACM Symposium on Eye Tracking Research and Applications (pp. 1-5).