Generative Adversarial Shaders for Real-Time Realism Enhancement

被引:0
|
作者
Salmi, A. [1 ]
Csefalvay, Sz [1 ]
Imber, J. [1 ]
机构
[1] Imaginat Technol, Kings Langley, England
关键词
Computing methodologies ? Computer graphics; Machine learning; Image processing; Rendering; Graphics processors;
D O I
10.1111/cgf.14870
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
Application of realism enhancement methods, particularly in real-time and resource-constrained settings, has been frustrated by the expense of existing methods. These achieve high quality results only at the cost of long runtimes and high bandwidth, memory, and power requirements. We present an efficient alternative: a high-performance, generative shader-based approach that adapts machine learning techniques to real-time applications, even in resource-constrained settings such as embedded and mobile GPUs. The proposed learnable shader pipeline comprises differentiable functions that can be trained in an end-to-end manner using an adversarial objective, allowing for faithful reproduction of the appearance of a target image set without manual tuning. The shader pipeline is optimized for highly efficient execution on the target device, providing temporally stable, faster-than-real time results with quality competitive with many neural network-based methods.
引用
收藏
页数:12
相关论文
共 50 条
  • [1] Real-time colouring and filtering with graphics shaders
    Vohl, D.
    Fluke, C. J.
    Barnes, D. G.
    Hassan, A. H.
    MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY, 2017, 471 (03) : 3323 - 3346
  • [2] Exploring real-time super-resolution generative adversarial networks
    Hu, Xiaoyan
    Wang, Zechen
    Liu, Xiangjun
    Li, Xinran
    Cheng, Guang
    Gong, Jian
    INTERNATIONAL JOURNAL OF SENSOR NETWORKS, 2021, 36 (02) : 85 - 96
  • [3] Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networks
    Li, Chuan
    Wand, Michael
    COMPUTER VISION - ECCV 2016, PT III, 2016, 9907 : 702 - 716
  • [4] Real-time segmentation of various insulators using generative adversarial networks
    Chang, Wenkai
    Yang, Guodong
    Yu, Junzhi
    Liang, Zize
    IET COMPUTER VISION, 2018, 12 (05) : 596 - 602
  • [5] RTSRGAN: Real-Time Super-Resolution Generative Adversarial Networks
    Hu, Xiaoyan
    Liu, Xiangjun
    Wang, Zechen
    Li, Xinran
    Peng, Wenqiang
    Cheng, Guang
    2019 SEVENTH INTERNATIONAL CONFERENCE ON ADVANCED CLOUD AND BIG DATA (CBD), 2019, : 321 - 326
  • [6] Management and control of shaders based on real-time rendering
    Chen, Yue-Hua
    Zheng, Chun-Hui
    Han, Ji-Qing
    Harbin Gongye Daxue Xuebao/Journal of Harbin Institute of Technology, 2009, 41 (03): : 63 - 66
  • [7] Physics-based generative adversarial network for real-time acoustic holography
    Lu, Qingyi
    Zhong, Chengxi
    Su, Hu
    Liu, Song
    ULTRASONICS, 2025, 149
  • [8] A REAL-TIME MEDICAL ULTRASOUND SIMULATOR BASED ON A GENERATIVE ADVERSARIAL NETWORK MODEL
    Peng, Bo
    Huang, Xing
    Wang, Shiyuan
    Jiang, Jingfeng
    2019 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2019, : 4629 - 4633
  • [9] Realistic real-time processing of anime portraits based on generative adversarial networks
    Zhu, Gaofeng
    Qu, Zhiguo
    Sun, Le
    Liu, Yuming
    Yang, Jianfeng
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2024, 21 (04)
  • [10] An Efficient Recurrent Adversarial Framework for Unsupervised Real-Time Video Enhancement
    Dario Fuoli
    Zhiwu Huang
    Danda Pani Paudel
    Luc Van Gool
    Radu Timofte
    International Journal of Computer Vision, 2023, 131 : 1042 - 1059