REU Site: Research Experience for Undergraduates in Computer Vision

这是计算机视觉本科生研究体验(REU)网站的延续，该网站在过去的36年里一直在中佛罗里达大学运营。每年，10名学生将参加为期10周的全日制暑期项目，沉浸在培养研究生的环境中。学员将接受培训，开始进行计算机视觉方面的研究；他们将特别学习计算机视觉的基础知识，计算机视觉在许多领域发挥着重要作用，包括安全、医疗保健、国防、商业和制造业。这个项目将为不同的本科生群体提供机会，他们大多来自以本科生为主的机构，与一个非常成功的研究小组度过一个夏天。参与者参加短期课程，将自己与他们最渴望的项目主题相匹配，并花费足够的时间进行重点研究。然后，他们可以选择与教授合作，为他们的项目撰写研究论文，为GRE做准备，并申请研究生课程，从而继续学习一年。我们的REU模式将在其他学校中传播。该项目包括开发新的计算机视觉算法，用于解决隐私保护活动和事件识别方面的科学问题，大型视觉语言建模，基础计算机视觉模型，图像和视频地理定位，视频对象分割，视频数据压缩，以及相关问题。这既涉及理论又涉及实践；我们对困难视觉问题的数学建模和分析以及开发算法感兴趣，同时构建在现实生活中演示这些解决方案的真实系统。这个项目将加强参与者的以下能力：编写复杂的计算机程序；就研究问题进行口头交流；思考现实生活中的计算机视觉问题；阅读和理解关键的研究论文；对计算机视觉问题的讨论做出贡献；以及考虑走上一条通往研究生职业生涯的道路。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This is a continuation of a Research Experience for Undergraduates (REU) site in Computer Vision which has operated at the University of Central Florida for the past thirty-six years. Each year, 10 students will participate in a 10-week duration, full-time Summer program, immersed in a nurturing graduate environment. Participants will be trained to begin to conduct research in Computer Vision; in particular, they will learn fundamentals of computer vision, which plays an important role in many areas, including, security, healthcare, defense, commerce and manufacturing. This project will provide opportunities to a diverse group of undergraduates, most from predominantly undergraduate institutions, to spend a summer with a highly successful research group. Participants take the short course, match themselves to a project topic that they most desire, and spend sufficient time in focused research. They then can opt to continue through the year by working with the professors to write research papers on their project, to prepare for the GREs and to apply to graduate programs. Our model for REU will be disseminated among other schools. This project includes development of new computer vision algorithms for solving scientific problems in privacy preserving activity and event recognition, Large Visual Language modeling, foundation computer vision models, image and video geo-localization, video object segmentation, data condensation in videos, and related problems. This involves both theory and practice; we are interested in mathematical modeling and analysis of difficult vision problems and developing algorithms, while at the same time building real systems for demonstrating those solutions in real life situations. This project will strengthen the following skills in a participant's ability to: write complex computer programs; communicate orally about research problem; think about computer vision issues in real life problems; read and understand key research papers; contribute to discussion of a computer vision problem; and, consider following a pathway towards a graduate career.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.