【复旦大学研发“眸思”大模型，为视障者打开新“视”界】
近日，复旦大学自然语言处理实验室（FudanNLP）传来一项重大科研突破，其团队成功研发出名为“眸思”（MouSi）的多模态大模型，旨在帮助视障人士“看见”世界。这一创新成果已转化为名为“听见世界”的App，为视障者的生活带来了革命性的改变。
据复旦大学官方公众号报道，这款App只需一部装有摄像头的设备和一对耳机，就能将实时拍摄的画面转化为语音描述，使视障用户能够通过听觉感知周围环境。更为先进的是，“眸思”系统不仅能够准确描述场景，如人物、物体和文字，还具备识别潜在风险并及时提示的功能，为视障人士的出行和生活提供了安全保障。
“眸思”大模型的诞生，是人工智能技术与社会公益事业的一次完美结合，体现了科技的人文关怀。该系统利用深度学习和自然语言处理技术，实现了视觉信息的高效转换，为视障者提供了更为独立、自主的生活方式。复旦大学的这一创新成果，无疑为全球视障人群带来了新的希望，也为无障碍科技领域树立了新的标杆。
这一消息在科技和慈善界引起了广泛关注，包括华尔街日报和纽约时报在内的多家国内外知名媒体对此进行了报道，高度赞扬了复旦大学团队的科研成就和社会责任感。此次“眸思”项目的成功，不仅展示了中国在人工智能领域的科研实力，也为全球无障碍科技发展提供了新的研究方向和实践案例。
英语如下：
**News Title:** “Fudan University Develops ‘MouSi’ Large Model: A New Hope for the Visually Impaired, Turning Camera Input into Voice,开创 ‘Hearing the World’ Era”
**Keywords:** MouSi Large Model, Visual Impairment Assistance, Fudan Research
**News Content:**
**Fudan University’s ‘MouSi’ Large Model Paves a New ‘Visual’ Path for the Visually Impaired**
Recently, the Fudan Natural Language Processing Laboratory (FudanNLP) announced a significant scientific breakthrough. Their team has successfully developed the “MouSi” (MouSi) multimodal large model, designed to help visually impaired individuals “see” the world. This innovative achievement has been transformed into an app called “Hearing the World,” revolutionizing the lives of the visually impaired.
According to the official Fudan University WeChat account, the app only requires a device with a camera and a pair of headphones to convert real-time images into voice descriptions, enabling visually impaired users to perceive their surroundings through sound. More impressively, the “MouSi” system not only accurately describes scenes, such as people, objects, and text but also has the capability to identify potential risks and provide timely alerts, ensuring safety for visually impaired individuals during travel and daily life.
The birth of the “MouSi” large model represents a perfect blend of artificial intelligence technology and social welfare, showcasing the humanistic concern of science. The system leverages deep learning and natural language processing technologies to efficiently convert visual information, offering visually impaired individuals a more independent and self-reliant lifestyle. Fudan University’s innovation has undoubtedly brought new hope to visually impaired people worldwide and set a new benchmark in the field of accessible technology.
This development has attracted widespread attention in the tech and philanthropic sectors, with multiple renowned domestic and international media outlets, including The Wall Street Journal and The New York Times, reporting on the achievement and commending the Fudan team’s research accomplishments and social responsibility. The success of the “MouSi” project not only demonstrates China’s research capabilities in artificial intelligence but also provides new research directions and practical cases for global accessible technology development.
【来源】https://www.ithome.com/0/753/295.htm
Views: 1