Method and apparatus for determining image to be labeled and model training method and apparatus

An apparatus for determining a target image, comprising a memory and a processor, the memory being configured to store a program;

• the processor being configured to execute the program in the memory to:
• obtain an original image and an autoencoder (AE) set, the original image being an image having not been labeled, the AE set comprising N AEs, N being an integer greater than or equal to 1,
• obtain an encoded image set corresponding to the original image by using the AE set, the encoded image set comprising N encoded images, the encoded images being corresponding to the AEs,
• obtain the encoded image set and a segmentation result set corresponding to the original image by using an image segmentation network, the image segmentation network comprising M image segmentation sub-networks, and the segmentation result set comprising [(N+1)*M] segmentation results, M being an integer greater than or equal to 1;
• determine a labeling uncertainty corresponding to the original image according to the segmentation result set obtained by the obtaining module;
• and
• determine whether the original image is a target image according to the labeling uncertainty.

A non-transitory computer-readable storage medium, comprising instructions, the instructions, when run on a computer, causing the computer to perform:

• obtaining an original image and an autoencoder (AE) set, the original image being an image having not been labeled, the AE set comprising N AEs, N being a positive integer;
• obtaining an encoded image set corresponding to the original image by using the AE set, the encoded image set comprising N encoded images, the encoded images being corresponding to the AEs;
• obtaining the encoded image set and a segmentation result set corresponding to the original image by using an image segmentation network, the image segmentation network comprising M image segmentation sub-networks, and the segmentation result set comprising [(N+1)*M] segmentation results, M being a positive integer;
• determining labeling uncertainty corresponding to the original image according to the segmentation result set;
• and
• determining whether the original image is a target image according to the labeling uncertainty.