Related Experiment Video
Updated: Jun 7, 2025
07:12
Development of a Gaze-Contingent Display Framework Designed for Perceptual and Oculomotor Research with Simulated Central Vision Loss
Published on: April 11, 2025
302
An audiovisual cognitive optimization strategy guided by salient object ranking for intelligent visual prothesis
Junling Liang1, Heng Li1, Xinyu Chai1
1School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200240, People's Republic of China.
Journal of Neural Engineering
|November 21, 2024
Summary
This study introduces an intelligent visual prosthesis system using AI to improve artificial vision. The system enhances object recognition and understanding for the blind, offering valuable insights for future prosthetic development.
Area of Science:
- Biomedical Engineering
- Artificial Intelligence
- Computer Vision
Background:
- Visual prostheses aim to restore sight but face real-world challenges.
- Advancements in AI and deep learning enable intelligent visual prosthetics with auditory support.
Purpose of the Study:
- To develop an AI-driven visual prosthesis system with auditory feedback for enhanced artificial vision.
- To improve object identification and spatial understanding for individuals with blindness.
Main Methods:
- An object-based attention mechanism simulating human gaze was developed.
- A salient object ranking (SaOR) network and dataset were created for depth perception.
- A SaOR-guided image description method provided auditory feedback, forming an audiovisual cognitive optimization strategy.
Main Results:
- The SaOR method improved object identification and correlation understanding in simulated prosthetic vision.
- The audiovisual cognitive optimization strategy further enhanced prosthetic visual cognition.
Conclusions:
- The developed system offers significant technical insights for next-generation intelligent visual prostheses.
- This research establishes a theoretical foundation for visual information processing strategies in prosthetic vision.
