Skip to main content
Fig. 6 | BioMedical Engineering OnLine

Fig. 6

From: Synthesizing multi-frame high-resolution fluorescein angiography images from retinal fundus images using generative adversarial networks

Fig. 6

Architecture of proposed HrGAN with common encoders \({G}_{e}\), decoders \({G}_{d1}\), \({G}_{d2}\), and \({G}_{d3}\), and discriminators \({D}_{1}\), \({D}_{2}\), and \({D}_{3}\). \({G}_{e}\), \({G}_{d1}\), and \({D}_{1}\) are used for transit-phase FA image generation, while \({G}_{e}\), \({G}_{d2}\), and \({D}_{2}\) are used for recirculation-phase FA image generation, and \({G}_{e}\), \({G}_{d3}\), and \({D}_{3}\) are used for transit-phase FA image generation

Back to article page