arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1208.0967cs.CV
收藏

使用来自RGB-D视频的对象示能性进行人类活动学习

Human Activity Learning using Object Affordances from RGB-D Videos

Cornell University(康奈尔大学)

Hema Swetha Koppula, Rudhir Gupta, Ashutosh Saxena

摘要:

人类活动包括按顺序执行的多个子活动,并涉及与各种对象的交互。这使得推理对象的启示活动识别的中心任务。在这项工作中,我们考虑的问题,共同标记的对象启示和人类活动的RGB-D视频。我们将问题框架为马尔可夫随机场,其中节点表示对象和子活动,边缘表示对象启示之间的关系,它们与子活动的关系以及它们随时间的演变。我们制定的学习问题,使用结构SVM的方法,其中标签在各种交替的时间分割被认为是潜在变量。我们在一个包含从四个主题收集的120个活动视频的数据集上测试了我们的方法,并获得了81.8%的端到端精确度和80.0%的召回率来标记活动。

英文摘要:

Human activities comprise several sub-activities performed in a sequence and involve interactions with various objects. This makes reasoning about the object affordances a central task for activity recognition. In this work, we consider the problem of jointly labeling the object affordances and human activities from RGB-D videos. We frame the problem as a Markov Random Field where the nodes represent objects and sub-activities, and the edges represent the relationships between object affordances, their relations with sub-activities, and their evolution over time. We formulate the learning problem using a structural SVM approach, where labeling over various alternate temporal segmentations are considered as latent variables. We tested our method on a dataset comprising 120 activity videos collected from four subjects, and obtained an end-to-end precision of 81.8% and recall of 80.0% for labeling the activities.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑