Discovering Object Classes from Activities

2014

Conference Paper

ps

In order to avoid an expensive manual labeling process or to learn object classes autonomously without human intervention, object discovery techniques have been proposed that extract visual similar objects from weakly labelled videos. However, the problem of discovering small or medium sized objects is largely unexplored. We observe that videos with activities involving human-object interactions can serve as weakly labelled data for such cases. Since neither object appearance nor motion is distinct enough to discover objects in these videos, we propose a framework that samples from a space of algorithms and their parameters to extract sequences of object proposals. Furthermore, we model similarity of objects based on appearance and functionality, which is derived from human and object motion. We show that functionality is an important cue for discovering objects from activities and demonstrate the generality of the model on three challenging RGB-D and RGB datasets.

@inproceedings{Srik:ECCV:2014,
  title = {Discovering Object Classes from Activities},
  author = {Srikantha, Abhilash and Gall, Juergen},
  booktitle = {European Conference on Computer Vision},
  volume = {8694},
  pages = {415-430},
  series = {Lecture Notes in Computer Science},
  editors = {D. Fleet  and T. Pajdla and B. Schiele  and T. Tuytelaars },
  publisher = {Springer International Publishing},
  month = sep,
  year = {2014},
  doi = {10.1007/978-3-319-10599-4_27},
  month_numeric = {9}
}