Bio-Inspired Visual Analytics

Humans are adept at complex visual tasks such as detection, tracking and recognition. The first aim of this project is to understand humans visual attention mechanism and the factors influencing it. In particular we investigate the importance of scene context in attention modeling. We have also developed human attention inspired models to build semantic priors for text detection which outperforms state-of-the-art. Additionally, we have also improved object detection and video object segmentation using eye tracking data.

Related publications

S. Karthikeyan, Thuyen Ngo, Miguel Eckstein and B.S. Manjunath,
“Eye tracking assisted extraction of attentionally important objects from videos”,
IEEE International Conference on Computer Vision and Pattern Recognition, Boston, MA, Jun. 2015.
[abstract] [PDF] [BibTex]
Abstract preview: Visual attention is a crucial indicator of the relative importance of objects in visual scenes to human viewers. In this paper, we propose an algorithm to extract objects which attract visual attentio ... more
S. Karthikeyan, Vignesh Jagadeesh, Renuka Shenoy, Miguel Eckstein and B.S. Manjunath,
“From Where and How to What We See”,
IEEE International Conference on Computer Vision, Sydney, Dec. 2013.
[abstract] [PDF] [BibTex]
Abstract preview: Eye movement studies have confirmed that overt attention is highly biased towards faces and text regions in images. In this paper we explore a novel problem of predicting face and text regions in imag ... more
S. Karthikeyan, Vignesh Jagadeesh and B. S. Manjunath,
“Learning Top Down Scene Context For Visual Attention Modelling In Natural Images”,
IEEE International Conference on Image Processing, Sep. 2013.
[abstract] [PDF] [BibTex]
Abstract preview: Top down image semantics play a major role in predicting where people look in images. Present state-of-the-art approaches to model human visual attention incorporate high level object detections signi ... more
S. Karthikeyan, Vignesh Jagadeesh and B. S. Manjunath,
“Learning Bottom-Up Text Attention Maps For Text Detection Using Stroke Width Transform”,
IEEE International Conference on Image Processing, Sep. 2013.
[abstract] [PDF] [BibTex]
Abstract preview: Humans have a remarkable ability to quickly discern regions containing text from other noisy regions in images. The primary contribution of this paper is to learn a model to mimic this behavior and ai ... more