CPC G06T 7/11 (2017.01) [G06N 3/045 (2023.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 20/41 (2022.01); G06V 20/695 (2022.01)] | 8 Claims |
1. A knowledge distillation based semantic image segmentation method, comprising:
inputting an input image to a teacher network and a student network;
normalizing a first feature vector corresponding to each pixel in a feature map of a last layer of the teacher network and normalizing a second feature vector corresponding to each pixel in a feature map of a last layer of the student network;
generating a first channel and space association matrix and a second channel and space association matrix based on the normalized first feature vector and the normalized second feature vector; and
defining a first loss function based on a Euclidean norm value of the difference between the first channel and space association matrix and the second channel and space association matrix,
wherein when the number of labels for the input image is C, the generating of a first channel and space association matrix includes:
calculating a first channel associated vector with respect to the C labels, based on the normalized first feature vector and a circularly shifted vector of the normalized first feature vector;
calculating a first channel association matrix by concatenating and realigning the first channel associated vector for the C labels; and
determining a first channel association matrix as a first channel and space association matrix.
|