VF-Lens: Enhancing Visual Perception of Visually Impaired Users in VR via Adversarial Learning with Visual Field Attention

This IEEE conference talk presents VF-Lens, a generative adversarial model designed to enhance image perception for visually impaired users in virtual reality environments. Learn how researchers from The University of Sydney, Beijing Technology and Business University, and Shandong University developed an adaptive system that compensates for light sensitivity and generates customized hyperimages based on users' visual field parameters. Discover how the innovative "visual field attention" mechanisms prioritize critical visual information, enabling visually impaired users to experience perceptions closer to those with normal vision. The presentation covers the model's applicability to various types of visual impairments while avoiding engineering complexities, along with the extensive evaluations that demonstrate its effectiveness and the standardized evaluation process ensuring reusability for future research in VR accessibility.