From Where and How to What We See

Karthikeyan, S.; Jagadeesh, Vignesh; Shenoy, Renuka; Ecksteinz, Miguel; Manjunath, B.S.

doi:10.1109/iccv.2013.83

Cited by 21 publications

(18 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This includes work on image segmentation, which shows that using fixation data to help segmentation algorithms leads to improved performance [32,37,48]. Eye tracking data is also useful for face and text detection: [25] cluster fixations to find regions in which the targets are likely to be found, and then apply standard face and text detectors only there. Several authors have collected eye tracking data for video and shown that saliency maps computed from these data can improve action recognition [31,46].…”

Section: Related Workmentioning

confidence: 99%

Training Object Class Detectors from Eye Tracking Data

Papadopoulos

Clarke

Keller

et al. 2014

Lecture Notes in Computer Science

104

View full text Add to dashboard Cite

Abstract. Training an object class detector typically requires a large set of images annotated with bounding-boxes, which is expensive and time consuming to create. We propose novel approach to annotate object locations which can substantially reduce annotation time. We first track the eye movements of annotators instructed to find the object and then propose a technique for deriving object bounding-boxes from these fixations. To validate our idea, we collected eye tracking data for the trainval part of 10 object classes of Pascal VOC 2012 (6,270 images, 5 observers). Our technique correctly produces bounding-boxes in 50% of the images, while reducing the total annotation time by factor 6.8× compared to drawing bounding-boxes. Any standard object class detector can be trained on the bounding-boxes predicted by our model. Our large scale eye tracking dataset is available at groups.inf.ed.ac.uk/calvin/eyetrackdataset/.

show abstract

Section: Related Workmentioning

confidence: 99%

Training Object Class Detectors from Eye Tracking Data

Papadopoulos

Clarke

Keller

et al. 2014

Lecture Notes in Computer Science

104

View full text Add to dashboard Cite

show abstract

“…For example, egocentric videos have been used to reconstruct 3D social gaze to analyze human-human interaction [15], and eye gaze information has been incorporated into first-person activity recognition [6]. The use of gaze information is also becoming more and more important in computer vision tasks such as object and scene recognition [11,20].…”

Section: Introductionmentioning

confidence: 99%

4th international workshop on pervasive eye tracking and mobile eye-based interaction

Pfeiffer

Stellmach

Sugano

2014

Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing: Adjunct Publication

View full text Add to dashboard Cite

Previous work on eye tracking and eye-based human-computer interfaces mainly concentrated on making use of the eyes in traditional desktop settings. With the recent growth of interest in smart glass devices and low-cost eye trackers, however, gaze-based techniques for mobile computing is becoming increasingly important. PETMEI 2014 focuses on the pervasive eye tracking paradigm as a trailblazer for mobile eye-based interaction and eye-based context-awareness. We want to stimulate and explore the creativity of these communities with respect to the implications, key research challenges, and new applications for pervasive eye tracking in ubiquitous computing. The long-term goal is to create a strong interdisciplinary research community linking these fields together and to establish the workshop as the premier forum for research on pervasive eye tracking.

show abstract

“…Utility of eye tracking data in action recognition techniques [42,29] have also shown promise. We note that the works in [33,25,36,43,45] are designed for object localization in images and in the proposed work we deal with object extraction in videos and therefore are not directly comparable as eye tracking data properties differ in images compared to videos. Additionally, eye tracking provides a platform to quickly annotate a large number of video frames which is of greater practical use compared to image annotation as manually annotating videos is a far more tedious task.…”

Section: Related Workmentioning

confidence: 99%

“…Recently there has been significant interest [30,36,43,33,42,29,25,31,44,45,46,38] in eye tracking enhanced computer vision. Relevant to our work, Mishra et al [30] proposed a segmentation using fixation approach which segments objects of interest given a single fixation point.…”

Section: Related Workmentioning

confidence: 99%

“…The final detection performance is considerably lower than perfect ground truth annotations, however these annotations are obtained in about a sixth of the time required to hand annotate the bounding boxes which is encouraging. In [25] we propose a technique to extract face and text semantic priors using eye tracking data from multiple subjects and use this to enhance stateof-the-art object detectors. The algorithm is designed for images and is targeted for only face and text categories.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation