Göm meny

Attentional Masking for Pre-trained Deep Networks

Marcus Wallenberg, Per-Erik Forssén
IROS17, Vancouver, Canada
Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS17)
September 2017


The ability to direct visual attention is a fundamental skill for seeing robots. Attention comes in two flavours: the gaze direction (overt attention) and attention to a specific part of the current field of view (covert attention), of which the latter is the focus of the present study. Specifically, we study the effects of attentional masking within pre-trained deep neural networks for the purpose of handling ambiguous scenes containing multiple objects. We investigate several variants of attentional masking on partially pre-trained deep neural networks and evaluate the effects on classification performance and sensitivity to attention mask errors in multi-object scenes. We find that a combined scheme consisting of multi-level masking and blending provides the best trade-off between classification accuracy and insensitivity to masking errors. This proposed approach is denoted multilayer continuous-valued convolutional feature masking (MC-CFM). For reasonably accurate masks it can suppress the influence of distracting objects and reach comparable classification performance to unmasked recognition in cases without distractors.

Bibtex entry

  author = 	 {Marcus Wallenberg and Per-Erik Forss\'en},
  title = 	 {Attentional Masking for Pre-trained Deep Networks},
  booktitle = {Proceedings of {IEEE/RSJ} International Conference on
Intelligent Robots and Systems ({IROS17})},
  year = 	 {2017},
  month = 	 {September},
  address = 	 {Vancouver, Canada},
  OPTorganization = {},
  publisher =    {{IEEE}}
  note = 	 {VR Project: Learnable Camera Motion Models, 2014-5928}

Per-Erik Forssén

Per-Erik Forssén


Computer Vision Laboratory
Department of Electrical Engineering
Building B
Room 2D:521
SE-581 83 Linköping, Sweden
+46(0)13 285654

< >

My pages:

Informationsansvarig: Per-Erik Forssén
Senast uppdaterad: 2017-08-10