• Sagar Vakkala
• Senthil Kumar
• Aniket Mishra
• Punitha D

Image super resolution has always had a great deal of proportion in the problem set solved by neural networks. The quality of the generated SR image has always been proportional to the time and resources put into the training of the network. With many methods like SR-GANs and Feedback loops for super resolution imaging all have their disadvantages, the biggest being the performance-time tradeoff. What SRFBGAN intends to do is use the best parts of both worlds, the high quality SR image provided by SR-GANs and the utilization of high resolution features by Feedback loops combined to produce a high quality SR image comparable to the other state-of-the-art architectures. This is accomplished by altering the structure of the Generator used to create the SR image. The novel generator architecture consists of feedback loops that ensure high level information is utilized when generating super resolution images.

SR:- Super Resolution, SISR:- Single Image Super Resolution, CNN: Convolution Neural Networks, GANs:- Generative Adversarial Networks, Feedback Loops, SRFBGAN:- Super Resolution Feedback Generative Adversarial Network

"SRFBGAN: Image super resolution using Feedback loops in Generative Adversarial Networks", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.7, Issue 5, page no.1-4, May-2020, Available :http://www.jetir.org/papers/JETIRDV06001.pdf

"SRFBGAN: Image super resolution using Feedback loops in Generative Adversarial Networks", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.7, Issue 5, page no. pp1-4, May-2020, Available at : http://www.jetir.org/papers/JETIRDV06001.pdf