Generative adversarial network has received widespread attention in the field of infrared and visible image fusion, but single path fusion is prone to loss of shallow information and limited fusion ability for branch feature extraction. This paper proposes a fusion method for infrared and visible images based on multi-path generative adversarial networks. In the generator, three input paths are constructed using the source images and the results of guided image filter to extract more source image feature information, in order to obtain detailed and rich fused images; Then, the convolutional layer adds a extract mask attention module to improve the efficiency of extracting significant information, introducing dense connections and residual connections, while improving the efficiency of feature transmission, it can obtain more important feature information of the source image. In the discriminator, dual discriminators are used to estimate the regional distribution of infrared and visible light images, avoiding the modal imbalance problem of losing contrast information in a single discriminator network. Through comparative experiments, it is shown that the proposed algorithm has achieved the best results on multiple objective evaluation indicators and has better visual effects.