Environment Exploration for Object-Based Visual Saliency Learning - Archive ouverte HAL Access content directly
Conference Papers Year : 2016

Environment Exploration for Object-Based Visual Saliency Learning

Abstract

Searching for objects in an indoor environment can be drastically improved if a task-specific visual saliency is available. We describe a method to incrementally learn such an object-based visual saliency directly on a robot, using an environment exploration mechanism. We first define saliency based on a geometrical criterion and use this definition to segment salient elements given an attentive but costly and restrictive observation of the environment. These elements are used to train a fast classifier that predicts salient objects given large-scale visual features. In order to get a better and faster learning, we use an exploration strategy based on intrinsic motivation to drive our attentive observation. Our approach has been tested on a robot in our lab as well as on publicly available RGB-D images sequences. We demonstrate that the approach outperforms several state-of-the-art methods in the case of indoor object detection and that the exploration strategy can drastically decrease the time required for learning saliency.
Fichier principal
Vignette du fichier
root.pdf (1.88 Mo) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-01289159 , version 1 (17-03-2016)

Identifiers

  • HAL Id : hal-01289159 , version 1

Cite

Céline Craye, David Filliat, Jean-François Goudou. Environment Exploration for Object-Based Visual Saliency Learning. International Conference on Robotics and Automation , May 2016, Stockholm, Sweden. ⟨hal-01289159⟩
222 View
512 Download

Share

Gmail Mastodon Facebook X LinkedIn More