Aiming at the defects of existing algorithms such as structural confusion and texture blurring in repairing complex murals, a dual-generation adversarial network model incorporating structural and textural feature guidance is proposed. Firstly, U-Net is introduced into the dual-generation network, and the texture and structure information extracted by using the direction and channel dual-attention mechanism guides the structure and texture decoders to complete the feature reconstruction of structure and texture respectively, and combines the void residual block and jump connection to achieve multi-scale feature fusion extraction. Secondly, the feature maps output from the two branches are deeply fused by the dual gated feature fusion module to complete the feature information interaction. Finally the restoration is completed through the joint discriminator confrontation. The experiment takes a non-national treasure real broken mural somewhere in Wutai Mountain as the object, and the results show that it exceeds the comparison method in several aspects, and the visual effect is clearer and more natural.