Researchers have introduced GazeFlow, a new framework designed to predict egocentric gaze trajectories by modeling gaze as a joint distribution of temporal positions. This model utilizes conditional flow matching (CFM) to learn a velocity field that transforms a Gaussian noise sample into a plausible gaze trajectory. GazeFlow conditions this velocity field on both visual features extracted from video and top-down task information, achieving state-of-the-art performance on standard datasets and demonstrating improved alignment with human gaze dynamics. AI
IMPACT This research advances egocentric gaze prediction, potentially improving applications that rely on understanding human visual attention.
RANK_REASON The cluster contains a research paper detailing a new framework for gaze prediction. [lever_c_demoted from research: ic=1 ai=1.0]
- alphaXiv
- arXiv
- CatalyzeX
- Conditional Flow Matching
- DagsHub
- GazeFlow
- Gotit.pub
- Hugging Face
- ScienceCast
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →