Limits...
Robust Eye Center Localization through Face Alignment and Invariant Isocentric Patterns.

Pang Z, Wei C, Teng D, Chen D, Tan H - PLoS ONE (2015)

Bottom Line: In addition, we apply our method to non-frontal faces to test its robustness and accuracy, which are essential in gaze estimation but have seldom been mentioned in previous works.Through extensive experimentation, we show that the proposed method can achieve a significant improvement in accuracy and robustness over state-of-the-art techniques, with our method ranking second in terms of accuracy.According to our implementation on a PC with a Xeon 2.5Ghz CPU, the frame rate of the eye tracking process can achieve 38 Hz.

View Article: PubMed Central - PubMed

Affiliation: School of Physics and Engineering, Sun Yat-Sen University, Guangzhou, China.

ABSTRACT
The localization of eye centers is a very useful cue for numerous applications like face recognition, facial expression recognition, and the early screening of neurological pathologies. Several methods relying on available light for accurate eye-center localization have been exploited. However, despite the considerable improvements that eye-center localization systems have undergone in recent years, only few of these developments deal with the challenges posed by the profile (non-frontal face). In this paper, we first use the explicit shape regression method to obtain the rough location of the eye centers. Because this method extracts global information from the human face, it is robust against any changes in the eye region. We exploit this robustness and utilize it as a constraint. To locate the eye centers accurately, we employ isophote curvature features, the accuracy of which has been demonstrated in a previous study. By applying these features, we obtain a series of eye-center locations which are candidates for the actual position of the eye-center. Among these locations, the estimated locations which minimize the reconstruction error between the two methods mentioned above are taken as the closest approximation for the eye centers locations. Therefore, we combine explicit shape regression and isophote curvature feature analysis to achieve robustness and accuracy, respectively. In practical experiments, we use BioID and FERET datasets to test our approach to obtaining an accurate eye-center location while retaining robustness against changes in scale and pose. In addition, we apply our method to non-frontal faces to test its robustness and accuracy, which are essential in gaze estimation but have seldom been mentioned in previous works. Through extensive experimentation, we show that the proposed method can achieve a significant improvement in accuracy and robustness over state-of-the-art techniques, with our method ranking second in terms of accuracy. According to our implementation on a PC with a Xeon 2.5Ghz CPU, the frame rate of the eye tracking process can achieve 38 Hz.

No MeSH data available.


Related in: MedlinePlus

Sample results of eye tracking on Talking Face Video, selected from the 1th, 11th, 30th, 50th and 100th sequences in Talking Face Video.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4591357&req=5

pone.0139098.g013: Sample results of eye tracking on Talking Face Video, selected from the 1th, 11th, 30th, 50th and 100th sequences in Talking Face Video.

Mentions: The Talking Face Video database is used here to evaluate the developed tracking system. The video contains 5000 frames which are taken from a person engaged in a conversation. Explicit shape regression with five initialization shapes is used to do a face alignment. Python and C++ are taken as the programming language to implement the algorithm in a single thread based on the Xeon 2.5Ghz. The average processing time for each frame with a size of 720*576 is 26ms. Taking e≤ 0.10 as the successfully detecting eye centers, a 96.2% accuracy is got in 5000 images sequences. In Fig 13, some selected results are shown. The testing results are made into a video, which is uploaded online as: http://v.youku.com/v_show/id_XMTMwODY0Nzg4OA==.html (also available as supporting information S1 Video).


Robust Eye Center Localization through Face Alignment and Invariant Isocentric Patterns.

Pang Z, Wei C, Teng D, Chen D, Tan H - PLoS ONE (2015)

Sample results of eye tracking on Talking Face Video, selected from the 1th, 11th, 30th, 50th and 100th sequences in Talking Face Video.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4591357&req=5

pone.0139098.g013: Sample results of eye tracking on Talking Face Video, selected from the 1th, 11th, 30th, 50th and 100th sequences in Talking Face Video.
Mentions: The Talking Face Video database is used here to evaluate the developed tracking system. The video contains 5000 frames which are taken from a person engaged in a conversation. Explicit shape regression with five initialization shapes is used to do a face alignment. Python and C++ are taken as the programming language to implement the algorithm in a single thread based on the Xeon 2.5Ghz. The average processing time for each frame with a size of 720*576 is 26ms. Taking e≤ 0.10 as the successfully detecting eye centers, a 96.2% accuracy is got in 5000 images sequences. In Fig 13, some selected results are shown. The testing results are made into a video, which is uploaded online as: http://v.youku.com/v_show/id_XMTMwODY0Nzg4OA==.html (also available as supporting information S1 Video).

Bottom Line: In addition, we apply our method to non-frontal faces to test its robustness and accuracy, which are essential in gaze estimation but have seldom been mentioned in previous works.Through extensive experimentation, we show that the proposed method can achieve a significant improvement in accuracy and robustness over state-of-the-art techniques, with our method ranking second in terms of accuracy.According to our implementation on a PC with a Xeon 2.5Ghz CPU, the frame rate of the eye tracking process can achieve 38 Hz.

View Article: PubMed Central - PubMed

Affiliation: School of Physics and Engineering, Sun Yat-Sen University, Guangzhou, China.

ABSTRACT
The localization of eye centers is a very useful cue for numerous applications like face recognition, facial expression recognition, and the early screening of neurological pathologies. Several methods relying on available light for accurate eye-center localization have been exploited. However, despite the considerable improvements that eye-center localization systems have undergone in recent years, only few of these developments deal with the challenges posed by the profile (non-frontal face). In this paper, we first use the explicit shape regression method to obtain the rough location of the eye centers. Because this method extracts global information from the human face, it is robust against any changes in the eye region. We exploit this robustness and utilize it as a constraint. To locate the eye centers accurately, we employ isophote curvature features, the accuracy of which has been demonstrated in a previous study. By applying these features, we obtain a series of eye-center locations which are candidates for the actual position of the eye-center. Among these locations, the estimated locations which minimize the reconstruction error between the two methods mentioned above are taken as the closest approximation for the eye centers locations. Therefore, we combine explicit shape regression and isophote curvature feature analysis to achieve robustness and accuracy, respectively. In practical experiments, we use BioID and FERET datasets to test our approach to obtaining an accurate eye-center location while retaining robustness against changes in scale and pose. In addition, we apply our method to non-frontal faces to test its robustness and accuracy, which are essential in gaze estimation but have seldom been mentioned in previous works. Through extensive experimentation, we show that the proposed method can achieve a significant improvement in accuracy and robustness over state-of-the-art techniques, with our method ranking second in terms of accuracy. According to our implementation on a PC with a Xeon 2.5Ghz CPU, the frame rate of the eye tracking process can achieve 38 Hz.

No MeSH data available.


Related in: MedlinePlus