1 / 8

Visual Attention

Visual Attention. More information in visual field than we can process at a given moment Solutions Shifts of Visual Attention related to eye movements Some targets “pop-out”. Visual Search. Feature Integration Theory (Ann Treisman) Parallel pre-attentive stage

topaz
Télécharger la présentation

Visual Attention

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Visual Attention • More information in visual field than we can process at a given moment • Solutions • Shifts of Visual Attention related to eye movements • Some targets “pop-out”

  2. Visual Search • Feature Integration Theory (Ann Treisman) • Parallel pre-attentive stage • Parallel feature detection modules (e.g. intensity, color, orientation) • Feature maps for each module • Feature maps combinations  Salience Map

  3. Attention Control • Two Mechanisms • Bottom-up • Biases the observer towards selecting stimuli based on their saliency • Importance of unexpected • Feature maps calculated based on differences to neighbors (human eye) • Top-down • directs the “spotlight of attention” under cognitive, volitional control • Importance of desired • Use of prior information (knowledge)

  4. Integration of Motion • Motion is of fundamental importance in biological vision systems • Contributes to visual attention as confirmed by Watanabe et al. • Weighted summation of the feature maps • Attending only the moving objects.

  5. Integration of Prior Knowledge (top-down) • Draw attention to regions with specific characteristics(e.g. human faces) • Generate a feature map based on skin activation • Link it with the rest of the maps

  6. Computational Model (Itti & Koch)

  7. Demonstrative Results Skin Map Intensity Map • Possible Regions of Interest: • Hands • Head • Shoulders Color Map Orientation Map Motion Map

  8. Itti & Koch map Equally Weighted maps Motion Map Motion & Skin biased May attend only the moving objects!

More Related