Goal:
Discover object-like key-segments in an unlabeled video, and learn appearance and shap models from them to automatically segment the foreground objects.
Tips:
1 : Intuitively, a region that moves differently from its surroundings and appears frequently throughout the video will likely be among the main objects of interest. Conversely, one that seldom occurs is more likely to be an uninteresting, background object.
本文旨在探讨如何在未标记的视频中发现类似目标的关键片段,并从中学习外观和形状模型,进而实现前景对象的自动分割。文章指出,那些运动方式与周围区域不同且在整个视频中频繁出现的区域更可能是主要的兴趣对象。
1764

被折叠的 条评论
为什么被折叠?



