A semantic-foveal Bayesian model that uses YOLOv5 detections and no human scanpath training generates target-present search fixations that outperform IVSN and approach gaze-trained models on several COCO-Search18 metrics.
Revisiting active per- ception,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Human Scanpath Prediction in Target-Present Visual Search with Semantic-Foveal Bayesian Attention
A semantic-foveal Bayesian model that uses YOLOv5 detections and no human scanpath training generates target-present search fixations that outperform IVSN and approach gaze-trained models on several COCO-Search18 metrics.