Header logo is am

Learning Where to Search Using Visual Attention

2016

Conference Paper

am


One of the central tasks for a household robot is searching for specific objects. It does not only require localizing the target object but also identifying promising search locations in the scene if the target is not immediately visible. As computation time and hardware resources are usually limited in robotics, it is desirable to avoid expensive visual processing steps that are exhaustively applied over the entire image. The human visual system can quickly select those image locations that have to be processed in detail for a given task. This allows us to cope with huge amounts of information and to efficiently deploy the limited capacities of our visual system. In this paper, we therefore propose to use human fixation data to train a top-down saliency model that predicts relevant image locations when searching for specific objects. We show that the learned model can successfully prune bounding box proposals without rejecting the ground truth object locations. In this aspect, the proposed model outperforms a model that is trained only on the ground truth segmentations of the target object instead of fixation data.

Author(s): Alina Kloss and Daniel Kappler and Hendrik P. A. Lensch and Martin V. Butz and Stefan Schaal and Jeannette Bohg
Book Title: Proceedings of the IEEE/RSJ Conference on Intelligent Robots and Systems
Year: 2016
Month: October
Publisher: {IEEE}

Department(s): Autonomous Motion
Research Project(s): Modeling Top-Down Saliency for Visual Object Search
Bibtex Type: Conference Paper (conference)
Paper Type: Conference

Event Name: IROS 2016
Event Place: Daejeon, Korea
State: Published
Attachments:

BibTex

@conference{kloss_iros_2016,
  title = {Learning Where to Search Using Visual Attention},
  author = {Kloss, Alina and Kappler, Daniel and Lensch, Hendrik P. A. and Butz, Martin V. and Schaal, Stefan and Bohg, Jeannette},
  booktitle = {Proceedings of the {IEEE/RSJ} Conference on Intelligent Robots and Systems}},
  publisher = {{IEEE}},
  month = oct,
  year = {2016},
  month_numeric = {10}
}