Abstract: A present disclosure is a method of segmenting an abnormal robust for complex autonomous driving scenes and a system thereof, specifically relates to the technical field of an image segmenting system. The system includes: a segmentation module, configured to transmit an obtained input image to the segmentation network to obtain a segmentation prediction image, and then quantify the uncertainty of a segmentation prediction by means of calculating two different discrete metrics; a synthesis module, configured to match a generated data distribution with a data distribution of the input image by utilizing a conditional generative adversarial network; a difference module, configured to model and calculate the input image, an generated image, the semantic feature map and the uncertainty feature map based on an encoder, a fusion module and a decoder, to generate the segmentation prediction images for the abnormal objects; a model training module; and an integrated prediction module.
Abstract: A present disclosure is a method of segmenting an abnormal robust for complex autonomous driving scenes and a system thereof, specifically relates to the technical field of an image segmenting system. The system includes: a segmentation module, configured to transmit an obtained input image to the segmentation network to obtain a segmentation prediction image, and then quantify the uncertainty of a segmentation prediction by means of calculating two different discrete metrics; a synthesis module, configured to match a generated data distribution with a data distribution of the input image by utilizing a conditional generative adversarial network; a difference module, configured to model and calculate the input image, an generated image, the semantic feature map and the uncertainty feature map based on an encoder, a fusion module and a decoder, to generate the segmentation prediction images for the abnormal objects; a model training module; and an integrated prediction module.