FLAME: Facial Landmark Heatmap Activated Multimodal Gaze Estimation

Person-independent models for the same lack precision due to anatomical differences of subjects, whereas person-specific calibrated techniques add strict constraints on scalability. To overcome these issues, we propose a novel technique, Facial Landmark Heatmap Activated Multimodal Gaze Estimation (FLAME), as a way of combining eye anatomical information using eye landmark heatmaps to obtain precise gaze estimation without Figure 1: Different types of gaze estimation methods: any person-specific calibration. Our evaluation demonstrates (a) person-independent technique, (b) person-specific technique, a competitive performance of about 10% improvement (c) FLAME. Training subjects are green and test on benchmark datasets ColumbiaGaze and EYEDIAP.