Similar to the generators, we have two discriminators, and . The role of the discriminator  is to discriminate between the real source image, , and the fake source image . We know that the fake source image is generated by the generator .

Given an image to discriminator , it returns the probability of the image being a real source image:

The following figure shows the discriminator , as you can observe it takes the real source image x and the fake source image generated by the generator F as inputs and returns the probability of the image being a real source image:

The role of discriminator is to discriminate between the real target image, , and the fake target image, . We know that fake target image is generated by the generator, . Given an image to discriminator , it returns the probability of the image being a real target image:

The following figure shows the discriminator , as you can observe it takes the real target image, and the fake target image generated by the generator, as inputs and returns the probability of the image being a real target image: