Abstract
This study analyzed the impact of NVIDIA's Deep Learning Super Sampling (DLSS) technology on the graphics performance of modern video games through a qualitative, descriptive documentary research approach. Six scientific articles on super-resolution, convolutional neural networks, temporal reconstruction, and data-driven rendering were reviewed to identify the theoretical foundations underlying DLSS. The findings indicate that this technology integrates advances in deep learning to reconstruct high-quality images from lower-resolution inputs, improving graphics performance while maintaining visual quality. Furthermore, DLSS shares key principles with models such as SRGAN and ESRGAN, particularly the use of spatial and temporal information to optimize image reconstruction. It is concluded that DLSS represents a practical application of scientific advances in super-resolution, providing an efficient solution for real-time graphics processing.References
Ledig C, Theis L, Huszár F, Caballero J, Cunningham A, Acosta A, et al. Photo-realistic single image super-resolution using a generative adversarial network (SRGAN). En: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2017. p. 4681-4690.
Wang X, Yu K, Wu S, Gu J, Liu Y, Dong C, et al. ESRGAN: Enhanced super-resolution generative adversarial networks. En: Proceedings of the European Conference on Computer Vision (ECCV) Workshops; 2018. p. 0-0.
Timofte R, Agustsson E, Van Gool L, Xu J, et al. NTIRE 2017 challenge on single image super-resolution: Methods and results. En: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW); 2017. p. 1110-1121.
Zhang J, Li S, Huang X, Chen Y. Advances in temporal super-resolution for real-time rendering. Comput Graph Forum. 2023;42(7):1-18.
Endo Y, Wetzstein G, Matsuoka T, Kelly P. Data-driven super resolution graphics. Santa Clara (CA): NVIDIA Research; 2023.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright (c) 2026 Yennifer Díaz Romero, Ángel Santiago Salazar Segura, Víctor Manuel Martínez García, Pedro Antonio Valdez Lizárraga (Author)