Method for eye tracking and eye tracking device

1 . An eye tracking device, comprising:

a transceiver; and

a processor, coupled to the transceiver, wherein the processor is configured to:

obtain an eye image set via the transceiver;

input the eye image set to a machine learning model to generate an estimated eye image, wherein the estimated eye image corresponds to a time point later than a time point of the eye image set;

perform interpolation between a first image of the eye image set and the estimated eye image to generate an interpolated eye image, wherein the interpolated eye image corresponds to a time point between a time point of the first image of the eye image set and the time point the estimated eye image; and

perform eye tracking according to the interpolated eye image and the estimated eye image.

2 . The eye tracking device according to claim 1 , wherein the eye image set comprises a first eye image corresponding to a first time point and a second eye image corresponding to a second time point different from the first time point.

3 . The eye tracking device according to claim 1 , wherein the processor is further configured to:

perform preprocessing on a historical eye image set before training the machine learning model according to the historical eye image set, wherein the preprocessing comprises translation, rotation, or shearing.

4 . The eye tracking device according to claim 1 , wherein the machine learning model comprises a neural network.

5 . A method for eye tracking, comprising:

obtaining an eye image set;

inputting the eye image set to a machine learning model to generate an estimated eye image, wherein the estimated eye image corresponds to a time point later than a time point of the eye image set;

performing interpolation between a first image of the eye image set and the estimated eye image to generate an interpolated eye image, wherein the interpolated eye image corresponds to a time point between a time point of the first image of the eye image set and the time point the estimated eye image; and

performing eye tracking according to the interpolated eye image and the estimated eye image.

6 . The method according to claim 5 , wherein the eye image set comprises a first eye image corresponding to a first time point and a second eye image corresponding to a second time point different from the first time point.

7 . The method according to claim 5 , further comprising:

performing preprocessing on a historical eye image set before training the machine learning model according to the historical eye image set, wherein the preprocessing comprises translation, rotation, or shearing.

8 . The method according to claim 5 , wherein the machine learning model comprises a neural network.