图 书 馆
把人类的注意力集中在“正确”的视觉数据上Focusing Human Attention on the "Right" Visual Data |
|
| 课程网址: | http://videolectures.net/icme2012_grauman_visual_data/ |
| 主讲教师: | Kristen Grauman |
| 开课单位: | 德克萨斯大学 |
| 开课时间: | 2012-09-18 |
| 课程语种: | 英语 |
| 中文简介: |  广泛的视觉传感器和前所未有的连通性使我们眼花wash乱，这些视觉数据来自在线照片集，家庭视频，新闻镜头，医学图像或监视提要。其中哪些图像和视频值得人们关注？本演讲的重点是在这个问题至关重要的两个问题设置上：对象类别的有监督的学习和无监督的视频摘要。在第一种设置中，挑战在于筛选候选训练图像，并选择那些如果被人标记将对识别系统最有帮助的图像。第二，挑战在于筛选一个长时间运行的视频，并仅选择摘要以供观众观看。我将介绍针对这些问题的最新研究，包括用于大规模主动学习的新颖算法和可穿戴式摄像机的以自我为中心的视频概要。这两个领域都展示了“半自动化”某些计算机视觉任务的重要性，并为大型视觉分析提出了令人兴奋的新应用。 |
| 课程简介: | Widespread visual sensors and unprecedented connectivity have left us awash with visual data‐‐‐from online photo collections, home videos, news footage, medical images, or surveillance feeds. Which images and videos among them warrant human attention? This talk focuses on two problem settings in which this question is critical: supervised learning of object categories, and unsupervised video summarization. In the first setting, the challenge is to sift through candidate training images and select those that, if labeled by a human, would be most informative to the recognition system. In the second, the challenge is to sift through a long‐running video and select only the essential parts needed to summarize it for a human viewer. I will present our recent research addressing these problems, including novel algorithms for large‐scale active learning and egocentric video synopses for wearable cameras. Both domains demonstrate the importance of "semi‐automating" certain computer vision tasks, and suggest exciting new applications for large‐scale visual analysis. |
| 关 键 词: | 视觉传感; 计算机视觉; 视频 |
| 课程来源: | 视频讲座网 |
| 数据采集: | 2020-11-12：zyk |
| 最后编审: | 2020-11-16：yxd |
| 阅读次数: | 35 |