RBPGAN: Recurrent Back-Projection GAN for Video Super Resolution
This addresses video quality enhancement for applications in computer vision, but it is incremental as it combines existing models.
The paper tackles video super resolution by proposing RBPGAN, which integrates RBPN and TecoGAN with Ping-Pong loss to generate temporally coherent and spatially detailed videos, outperforming earlier work as demonstrated on datasets.
Recently, video super resolution (VSR) has become a very impactful task in the area of Computer Vision due to its various applications. In this paper, we propose Recurrent Back-Projection Generative Adversarial Network (RBPGAN) for VSR in an attempt to generate temporally coherent solutions while preserving spatial details. RBPGAN integrates two state-of-the-art models to get the best in both worlds without compromising the accuracy of produced video. The generator of the model is inspired by RBPN system, while the discriminator is inspired by TecoGAN. We also utilize Ping-Pong loss to increase temporal consistency over time. Our contribution together results in a model that outperforms earlier work in terms of temporally consistent details, as we will demonstrate qualitatively and quantitatively using different datasets.