New framework EgoCogNav models human cognition for egocentric navigation

Researchers have introduced EgoCogNav, a new multimodal framework designed to improve egocentric navigation by incorporating human cognitive and experiential factors. This framework jointly forecasts perceived path uncertainty, trajectories, and head motion using egocentric video, gaze, and motion history. To support further research, the team also released the Cognition-aware Egocentric Navigation (CEN) dataset, which includes six hours of real-world egocentric recordings capturing diverse navigation behaviors. AI

RANK_REASON The cluster describes a new research paper and dataset release on arXiv. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• arXiv
• Connected Papers
• CORE Recommender
• EgoCogNav
• Hugging Face
• IArxiv Recommender
• Litmaps
• scite Smart Citations
• Zhiwen Qiu

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →