NLOOK: a computational attention model for robot vision

Milton Roberto Heinen,Paulo Martins Engel

doi:10.1007/bf03194502

Abstract

Abstract The computational models of visual attention, originally proposed as cognitive models of human attention, nowadays are being used as front-ends to some robotic vision systems, like automatic object recognition and landmark detection. However, these kinds of applications have different requirements from those originally proposed. More specifically, a robotic vision system must be relatively insensitive to 2D similarity transforms of the image, as in-plane translations, rotations, reflections and scales, and it should also select fixation points in scale as well as position. In this paper a new visual attention model, called NLOOK, is proposed. This model is validated through several experiments, which show that it is less sensitive to 2D similarity transforms than other two well known and publicly available visual attention models: NVT and SAFE. Besides, NLOOK can select more accurate fixations than other attention models, and it can select the scales of fixations, too. Thus, the proposed model is a good tool to be used in robot vision systems.

Highlights

The amount of information coming down the optic nerve in the primate’s visual system, estimated to be on the order of 108 bits per second, far exceeds what the brain is capable of fully processing and assimilating into conscious experience[35]
This paper presents a new visual attention model, called NLOOK, which was intended to be used in robotic vision systems
NLOOK: Mean = 49.15; Standard dev. = 15.59. These results show that SAFE fixations are really more redundant than NLOOK fixations, i.e., the focus of attention (FOA) selected by SAFE are more overlapped than the FOAs selected by NLOOK

Summary

Introduction

The amount of information coming down the optic nerve in the primate’s visual system, estimated to be on the order of 108 bits per second, far exceeds what the brain is capable of fully processing and assimilating into conscious experience[35]. According to Desimone and Duncan[6], the interest regions selection is driven by a competitive attention control mechanism, which facilitates the emergency of a winner among several potential targets. This mechanism allows the visual system to process relevant information to current tasks, while suppressing the irrelevant information that cannot be analyzed simultaneously[16]. The visual attention, together with other mechanisms, allows the human being to have a wide vision field and an accurate detail perception without exceeding the capacity to consciously assimilate it[35]

Methods

Results

Conclusion