Gaze correction with a single webcam

Eye contact is a critical aspect of human communication. However, when talking over a video conferencing system, such as Skype, it is not possible for users to have eye contact when looking at the conversation partner's face displayed on the screen. This is due to the location disparity between the video conferencing window and the camera. This issue has been tackled by expensive high-end systems or hybrid depth+color cameras, but such equipment is still largely unavailable at the consumer level and on platforms such as laptops or tablets. In contrast, we propose a gaze correction method that needs just a single webcam. We apply recent shape deformation techniques to generate a 3D face model that matches the user's face. We then render a gaze-corrected version of this face model and seamlessly insert it into the original image. Experiments on real data and various platforms confirm the validity of the approach and demonstrate that the visual quality of our results is at least equivalent to those obtained by state-of-the-art methods requiring additional equipment.

[1]  Milton Chen,et al.  Leveraging the asymmetric sensitivity of eye contact for videoconference , 2002, CHI.

[2]  Olga Sorkine,et al.  Laplacian Mesh Processing , 2005 .

[3]  M. Argyle,et al.  Gaze and Mutual Gaze , 1994, British Journal of Psychiatry.

[4]  Yutaka Matsushita,et al.  Multiparty videoconferencing at virtual social distance: MAJIC design , 1994, CSCW '94.

[5]  Jörn Ostermann,et al.  3D talking head customization by adapting a generic model to one uncalibrated picture , 2001, ISCAS 2001. The 2001 IEEE International Symposium on Circuits and Systems (Cat. No.01CH37196).

[6]  Andrew Blake,et al.  Gaze manipulation for one-to-one teleconferencing , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[7]  Thomas Vetter,et al.  A morphable model for the synthesis of 3D faces , 1999, SIGGRAPH.

[8]  Andrew Jones,et al.  Achieving eye contact in a one-to-many 3D video teleconferencing system , 2009, ACM Trans. Graph..

[9]  Luc Van Gool,et al.  Blue-c: a spatially immersive display and 3D video portal for telepresence , 2003, IPT/EGVE.

[10]  John F. Canny,et al.  MultiView: spatially faithful group video conferencing , 2005, CHI.

[11]  Philippe Bekaert,et al.  Optimized Two-Party Video Chat with Restored Eye Contact Using Graphics Hardware , 2008, ICETE.

[12]  Tat-Jen Cham,et al.  Analogous view transfer for gaze correction in video sequences , 2002, 7th International Conference on Control, Automation, Robotics and Vision, 2002. ICARCV 2002..

[13]  Ruigang Yang,et al.  Eye gaze correction with stereovision for video-teleconferencing , 2002, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Wen Gao,et al.  Efficient 3D reconstruction for face recognition , 2005, Pattern Recognit..

[15]  Ruigang Yang,et al.  Eye contact in video conference via fusion of time-of-flight depth sensor and stereo , 2011 .

[16]  A. Monk,et al.  A Look Is Worth a Thousand Words: Full Gaze Awareness in Video-Mediated Conversation , 2002 .

[17]  Markus H. Gross,et al.  Gaze Correction for Home Video Conferencing , 2012 .

[18]  Simon Lucey,et al.  Deformable Model Fitting by Regularized Landmark Mean-Shift , 2010, International Journal of Computer Vision.

[19]  Kentaro Toyama,et al.  Gaze Awareness for Video-Conferencing: A Software Approach , 2000, IEEE Multim..

[20]  Hiroshi Ishii,et al.  ClearBoard: a seamless medium for shared drawing and conversation with eye contact , 1992, CHI.

[21]  C. N. Macrae,et al.  Are You Looking at Me? Eye Gaze and Person Perception , 2002, Psychological science.

[22]  Jim Gemmell,et al.  Implementing Gaze-corrected Videoconferencing , 2002, Communications, Internet, and Information Technology.