Method and device for determining parameter for gaze tracking device

A device for determining parameter for a gaze tracking device, comprising at least one processor;

• and a computer readable storage, coupled to the at least one processor and storing at least one computer executable instructions thereon, which when the at least one computer executable instructions is executed by the at least one processor, cause the at least one processor to carry out following actions:
• acquiring an image to be detected and a characteristic matrix of the image to be detected;
• processing the image to be detected according to a preset model to obtain a first vector and second vector of the image to be detected;
• and
• determining a parameter of the image to be detected according to the first vector and the second vector, the parameter of the image to be detected comprising at least one of:
• a position of a glint in the image to be detected and a gaze direction of eyes in the image to be detected;
• acquiring multiple images to be trained;
• processing the multiple images to be trained to obtain a characteristic matrix of the multiple images to be trained;
• and training the characteristic matrix of the multiple images to be trained to obtain third vectors, fourth vectors and an correlation matrix;
• wherein training the characteristic matrix of the multiple images to be trained to obtain the third vectors, the fourth vectors and the correlation matrix comprises:
• decomposing the characteristic matrix of the multiple images to be trained to obtain the third vectors;
• transforming the characteristic matrix of the multiple images to be trained according to the third vectors to obtain a first characteristic matrix;
• decomposing the first characteristic matrix to obtain the fourth vectors;
• transforming the first characteristic matrix according to the fourth vectors to obtain a second characteristic matrix;
• and under the condition that both of the third vectors and the fourth vectors meet a preset condition, obtain the correlation matrix according to the third vectors and the fourth vectors.

A non-transitory storage medium, comprising a stored program, the program executing the following instruction:

• acquiring an image to be detected and a characteristic matrix of the image to be detected;
• processing the image to be detected according to a preset model to obtain a first vector and second vector of the image to be detected;
• and
• determining a parameter of the image to be detected according to the first vector and the second vector, the parameter of the image to be detected comprising at least one of:
• a position of a glint in the image to be detected and a gaze direction of eyes in the image to be detected;
• the program before processing the image to be detected according to the preset model, further comprising:
• acquiring multiple images to be trained;
• processing the multiple images to be trained to obtain a characteristic matrix of the multiple images to be trained;
• and training the characteristic matrix of the multiple images to be trained to obtain third vectors, fourth vectors and a correlation matrix;
• wherein training the characteristic matrix of the multiple images to be trained to obtain the third vectors, the fourth vectors and the correlation matrix comprises:
• decomposing the characteristic matrix of the multiple images to be trained to obtain the third vectors;
• transforming the characteristic matrix of the multiple images to be trained according to the third vectors to obtain a first characteristic matrix;
• decomposing the first characteristic matrix to obtain the fourth vectors;
• transforming the first characteristic matrix according to the fourth vectors to obtain a second characteristic matrix;
• and under the condition that both of the third vectors and the fourth vectors meet a preset condition, obtaining the correlation matrix according to the third vectors and the fourth vectors.