Graffiti-based weak supervision semantic segmentation method and system

Abstract

The invention provides a graffiti-based weak supervision semantic segmentation method and system, and the method comprises the steps: obtaining a plurality of training images which correspond to graffiti marks and edge images; selecting a training picture as a current picture, and inputting the current picture into the semantic segmentation network to obtain high-level semantic features of the current picture; inputting the high-level semantic features into a prediction correction network to obtain a segmentation result graph of the current picture, and obtaining cross entropy loss of a graffiti mark area in the current picture according to the graffiti mark of the current picture; inputting the high-level semantic features into a boundary regression network to obtain a boundary map of a target in the current picture, and obtaining mean variance loss of a boundary region in the boundary map according to an edge map of the current picture; constructing a total loss function, judging whether the total loss function is converged or not, and if yes, taking the current prediction correction network as a semantic segmentation model; and inputting the picture to be semantically segmentedinto the semantic segmentation model to obtain a segmentation result picture of the picture to be semantically segmented.

Description

technical field [0001] The method belongs to the field of machine learning and computer vision, and in particular relates to machine learning problems for weakly supervised semantic segmentation in computer vision. Background technique [0002] The current popular scene segmentation methods are mainly based on the Fully Convolutional Network (FCN) and its variants. These methods all combine the idea of ​​transfer learning, using the pre-trained convolutional neural network on a large-scale image classification dataset, adjusting it to a fully convolutional network structure and retraining on a weakly supervised semantic segmentation dataset. For fine-label training, this method can achieve good segmentation results. However, when only weakly labeled training networks are provided, such methods mainly suffer from the following problems: (1) Inconsistencies and discontinuities often appear in the segmentation results, and (2) the segmentation boundaries of objects are often i...

AI Technical Summary

Problems solved by technology

However, when only weakly labeled training networks are provided, such methods mainly suffer from the following problems: (1) Inconsistencies and discontinuities often appear in the segmentation results, and (2) the segmentation boundaries of objects are often imprecise and incoherent

The segmentation model trained directly with graffiti marks can only produce rough segmentation results, mainly because the graffiti marks only contain part of the semantic information and do not provide fine boundary information to guide the model to accurately segment each target

Images