Ex-Jobb Proposal: Estimation of gaze direction
ثبت نشده
چکیده
The aim of this project is to estimate the gaze direction of a football player from video footage. The data processed and analyzed will be video obtained from professional football games, specifically from the “Allsvenskan” league. Currently, the Computer Vision and Perception Lab (CVAP) here at KTH is in collaboration with the Stockholm based, digital media and sports analysis company TRACAB. The latter has a real time player tracking system that estimates each player’s position on the football pitch throughout a game. It is employed regularly on “Allsvenskan” and “Champions League” games (buy a copy of Aftonbladet or check out http://zoom.sportbladet.se/ to see their results). The collaboration between CVAP and TRACAB focuses on building a 3D reconstruction of each player’s motion from the tracking and video data. This ex-jobb project would be one element of this whole process, see http: //www.csc.kth.se/∼sullivan/actvis. The raw data the student will obtain is a sequence of images containing one player. There will then be two distinct objectives. The first will be the construction of a head detector which will be used to localize the player’s head in each frame. It is envisaged that this head detector will be based on computer vision techniques which have proven to be successful in similar tasks [1]. Once the head has been localized the next step will be to estimate the direction the head is pointing. One initial approach to try will be that described in [2]. Simply, it involves extracting a feature vector based on counting the number of skin pixels in the head region and comparing this to feature vectors (labelled training data) of the some form which were extracted from images in which the head direction was known. It is then assumed if the newly extracted feature vector is similar to that of one of the labelled training examples then their corresponding gaze directions will also be similar. It is unclear at this stage whether this approach will be sufficient for the problem at hand, but some improvement may be required.
منابع مشابه
The effect of adding gaze direction recognition to stabilizing exercises on pain, muscular endurance and proprioception women with chronic non-specific neck pain
Aims and background: Gaze direction recognition is one of the new treatments method for neck pain. The positive effects of stabilization exercises in various studies on neck pain have also been confirmed. Therefore, the aim of this study was to investigate the effect of adding a gaze direction recognition program to common stabilizing exercises on neck pain intensity, muscular endurance and pro...
متن کاملA Method of Gaze Direction Estimation Considering Head Posture
A new method for head posture and gaze direction estimation is proposed. Firstly, three models of head position are established and postures are judged based on triangle attribute constituted by eyes and mouth. Then pupil is located using Hough transform in eye area. With the method of horizontal and vertical projection and eye prior knowledge, the normal eye outline is fitted. Finally, gaze di...
متن کاملA Study on Gaze Estimation Using Head and Body Pose Information
Gaze estimation from an image is an important technique for tasks such as driver monitoring and measuring of advertising effectiveness. For this, most existing methods require a high quality image of eyes. However, It is difficult to obtain the eye images when eye occlusion occurs due to sunglasses or face rotation. Another approach approximates head directions to gaze directions. However, the ...
متن کاملOMEG: Oulu Multi-Pose Eye Gaze Dataset
Data is in a very important position for pattern recognition tasks including eye gaze estimation. In the literature, most researchers used normal face datasets, which are not specifically designed for eye gaze estimation. As a result, it is difficult to obtain fine labeled eye gaze direction. Therefore large datasets with well-defined gaze directions are desired. To facilitate related researche...
متن کاملHuman-computer Interaction Using Head Movements
This paper presents a real-time gaze direction estimation approach. The face region and facial features are extracted automatically by applying skin detection, gray-scale morphology and a geometrical face model. The Kalman filtering is applied for tracking and estimation of the 3-D pose of the moving head. The rough gaze direction is determined from the orientation of the head. Experiments with...
متن کامل