Skip to main content


Fig. 4 | EURASIP Journal on Image and Video Processing

Fig. 4

From: A review of image-based automatic facial landmark identification techniques

Fig. 4

Ground truth variation. a Ground truth landmarks provided by individual annotators. Landmarks in red are the individual positions provided by the annotators, while the yellow points indicate the mean position for each of the landmarks. b Ground truth landmarks variation. Points in red are the mean positions for each landmark, while the major and minor axes of the blue ellipses indicate the x and y variance of the ground truth positions

Back to article page